Acceptance into a doctoral program is only the first step in getting a degree; the next pressing issue is how to pay for it. Although many students receive funding from their graduate department or their supervisor's grant, some must find other sources of financial support. Even if a student receives a tuition waiver and a stipend, it isn't always enough to live on, especially with increasing housing costs and perhaps a family to take care of.

So, how do you pay for graduate school? Students have to entertain a variety of questions: What kinds of funding are available? How long will it last? Are their work requirements? MiSciNet talks to several Ph.D. students and faculty members to find answers to these questions.

The bottom line: Through a combination of research assistantships (RAs), teaching assistantships (TAs), fellowships, and other means, almost all Ph.D. students in the sciences pay for their educations and support themselves--if not always in grand style--without incurring any educational debt. And the few who do incur some debt should think of their educations as an investment.

Learn Your Ships

The majority of doctoral students support themselves with TAs, fellowships, and RAs.

Statistically, TAs account for about a fifth--20.5%--of graduate students' primary support. For mathematics students, TAs are the primary means of support,according to a recent National Science Foundation (NSF) report, Science and Engineering Indicators 2006 . Many students have TAs early in their graduate careers, while they are still taking classes and before they've started on their thesis projects. TAs require recipients to teach or assist with undergraduate classes, which are usually, but not always, laboratory courses. Apart from the actual teaching, TA obligations include grading papers and meeting students during office hours to answer questions. Some TAs teach "recitation" sessions, which usually take the form of supplementary lectures or Q&A sessions associated with larger lecture courses.

The least common funding source is also the most prestigious: 9.5% of graduate students are supported mainly by fellowships. Fellowships usually come from outside the university--the federal government and science-related nonprofits are the most common sources--although most universities offer a few graduate fellowships. Students with fellowships have fewer nonresearch obligations--they don't have to teach or do other work that's not directly related to their education--so they can focus on their studies during the early years of their program and on their theses later.

"It's nice when a student brings in their own money ... because they don't have to TA or they don't have to TA as much," says Elma Gonzalez of the department of ecology and evolutionary biology at the University of California, Los Angeles (UCLA). And with resources independent of the university, the student can choose a thesis project without having to worry about their supervisor's funding situation, assuming the fellowship lasts long enough.

Fellowships are competitive--often very competitive. Emilio Bruna, an assistant professor of ecology at the University of Florida (UF), Gainesville, encourages students accepted to his lab to apply for national fellowships before arriving and helps those who don't have any experience writing fellowship applications by conducting workshops on the subject. Gonzalez's department, too, offers training in applying for fellowships, but most students are largely on their own in this area. "My feeling," says Richard Weiss, a biochemistry professor at UCLA, "is that it should be more proactive than it actually is."

RAs are the most common source of graduate-student support in the sciences, accounting in 2003 for 31.6% of graduate students' support, according to the NSF report Graduate Students and Postdoctorates in Science and Engineering: Fall 2003 . The money to pay for a research assistantship, says Bruna, comes from a research grant awarded to the graduate student's research supervisor (a.k.a. the principal investigator, or PI). A student receives an RA to work on a specific project--usually, but not always, the student's thesis project. Bruna provides RAs for his research students with an NSF-funded $16,000 stipend. "I won't take on a graduate student unless I can support them in some way," says Bruna. In return, Bruna's students must commit 10 hours each week to research. Generally, however, once they've finished classes and begun work on their thesis projects, most RAs work full time--or more--on research.

Looking for the Best Deal

Almost all graduate students receive a stipend, but the amount varies widely. "There's a lot of variation in terms of how much support students get," Bruna says. There are even differences, he notes, between departments at the same school. Incoming doctoral candidates at UF's Department of Chemistry receive $18,500 per year, whereas those entering the university's Biochemistry and Molecular Biology Department are offered $22,000.Stipend offers also depend on an area's cost of living and the amount offered at comparable departments.

Looking for Funding?

GrantsNet, sponsored by AAAS, is a Web site that allows students to search either an undergraduate or a graduate database of funding opportunities. Hundreds of fellowships, scholarships, and internships are available through this free service.

And, although most students have some kind of financial support, money isn't always available at every institution. Kenneth Byrd, a Ph.D. engineering student in Washington, D.C., was accepted into several programs that didn't offer financial support. He declined these, as well as one offer that, he says, did not provide enough money to live in Washington, D.C., where the university was located. Eventually, he was offered admission and funding from two other colleges. University A offered more money, but with a strict teaching requirement. University B offered a university fellowship with no service obligations, but the stipend was smaller. Byrd convinced the engineering department at University B to make up the difference and pay his administrative fees, with support guaranteed for 3 years. "I needed to know that I would not run out of money for my Ph.D.," he says.

PIs with grants from the National Institutes of Health (NIH) or, in certain engineering fields, from NSF, can apply for research supplements to support minority students. These supplements allow faculty members to provide support to minority students and postdocs who want to do research in that lab. For more information, see NIH's Research Supplements to Promote Diversity in Health-Related Research and NSF's Graduate Research Supplements programs.

The Cost of Living

Getting enough money to cover living expenses is a concern for graduate students, especially in areas with a high cost of living. Gonzalez, Bruna, and Weiss agree that doctoral stipends generally provide a livable wage, but only for those willing to accept what Weiss refers to as a "modestly lower-middle-class existence." Weiss suggests that graduate students live together and pool their resources, but even so, he says, at $23,000 annually--the average stipend for biomedical students at UCLA--most students can't afford half the $1600 average cost of a two-bedroom apartment close to the UCLA campus.

Consequently, says Barbara Ige, program director of UCLA's NSF Alliance for Graduate Education and the Professoriate (AGEP) in the Science, Technology, Engineering, and Mathematics program, many UCLA AGEP students choose to commute an hour or more each way. There are alternatives, however. Gonzalez recommends becoming an RA so that the university pays for oncampus housing. And, Ige says, some UCLA departments are able to guarantee graduate housing for 2 years for new students. Furthermore, many universities in urban areas subsidize housing. Columbia University and NYU, for example, have vast real-estate holdings in New York City and offer their graduate students, postdocs, and faculty members housing at below-market rates.

Combining Sources

Most students are supported, at various times, by two or more of the three major funding sources. A good example is Kobi Abayomi, a fourth-year Ph.D. student in statistics at Columbia, who paid for school with a combination of RAs and fellowships. The Center for International Earth Science Information Network gave him an RA to cover his first year, and the next 2 years he received an NSF Integrative Graduate Education and Research Traineeship fellowship. For his fourth and fifth years, the statistics department is providing a university fellowship. "I chose Columbia because of the strength of their program, not [because of] the level of funding," he says. Still, "I've been lucky to have some of my funding attached to my graduate work. But, in general, I think it is preferable to seek outside funding."

Graduate Debt

Most Ph.D. students manage to support themselves, however modestly, with a combination of TAs, RAs, and fellowships. But "if they've participated in master's programs," Ige says, "most likely they've incurred debt." But not to worry: Luis Mena, a lecturer for the Latino Financial Issues Program at the University of Texas, believes that this kind of debt shouldn't be a major concern for aspiring scientists. He argues that the investment in a graduate education is "good debt" because it reaps long-term financial benefits for the investor, unlike buying a luxury car or a home entertainment system, both of which depreciate over time. Like any investment, he says, knowing the ins and outs--the kinds of funding, the duration, and the requirements--is the best way to insure the best return on your investment.

Comments, suggestions? Please send your feedback to our editor .

Clinton Parks is a writer for Minority Scientists Network.

Clinton Parks is a writer for MiSciNet and may be reached at cparks@aaas.org.