NSF Fellowships Weather Politics, Promote Superb Science

WASHINGTON -- Retiring Massachusetts Institute of Technology president Paul Gray was one. So were Stanford University president Donald Kennedy and Harvard University paleontologist and author Stephen Jay Gould, not to mention Nobel laureates Thomas Cech, Walter Gilbert, and Burton Richter. Like many others among the nation's scientific leaders, these researchers got their starts as National Science Foundation Fellows. Each year NSF gives out hundreds of three-year awards to the nation's outstanding graduate students. The program was set up in 1952 to maintain the pace of research set during World War II. Last year, the program offered its 25,000th award, and foundation officials marked the event by pausing to see what some of the fellows have accomplished. The longevity of the program is itself a cause for celebration. "That's no small achievement in a political system with continuous pressure to make and remake and rename programs," says Terence Porter, director of research career development at NSF. The program's achievements are indeed impressive. The first group of 569 fellows named in 1952 included Richter, now director of the Stanford Linear Accelerator Center, and James W. Cronin, now at the University of Chicago, both winners of the Nobel Prize in Physics. Biologists Edward O. Wilson, Robert Selander, and Richard Lewontin were also part of that inaugural class. Notable women from that year include Joan Berkowitz, a physical chemist who in 1983 was recognized by the Society of Women Engineers for her accomplishments, and Maxine Singer, a biochemist who now heads the Carnegie Institution in Washington, D.C., and whose daughter, Stephanie, has also been an NSF fellow. In all, NSF counts among the fellows selected a dozen Nobelists, most recently University of Colorado chemist Thomas Cech. More than 300 former fellows belong to the National Academy of Sciences or the Academy of Engineering. Seven received NSF's Waterman Award for outstanding achievement before the age of 35; six have garnered the prestigious Fields Medal in mathematics. A total of 25 have been named five-year MacArthur Foundation "geniuses," and 10 are university presidents. One - geologist and former United States Republican Senator Harrison Schmitt of New Mexico - even went to the moon. Studies of their achievements done in 1977 and in 1988 show that NSF fellows take about 30 percent less time to finish their degrees than do other graduate students. More of them plan to do postdoctoral work as well. They also publish more papers: Fellows from 1960 to 1964, for example, had 40 percent more articles to their credit than the average for all Ph.D.'s, and were cited more than twice as often. Several other federal agencies and private organizations have set up programs modeled after the NSF fellowships. But none can match the NSF program's reputation. "The NSF fellows are consistently the intellectual leaders in every graduate school cohort," says John H. D'Arms, dean of the Horace H. Rackham School of Graduate Studies at the University of Michigan. In many ways, the fellows have changed little since the 1950s. Take Robert Hanson, a professor of chemistry at St. Olaf College in Northfield, Minn. Hanson, 32, grew up in a small Colorado town and became interested in science as a result of an NSF-sponsored summer program for high school students. After his graduation from Caltech, he used his NSF fellowship to head east for study at Columbia University, when he remembers walking down the hall and saying, "This is so great . . . and I'm being paid for it." The terms of the fellowship have evolved over the years to meet the changing needs of students and to stretch limited funds. What began as one-year, renewable stipends available primarily to advanced graduate students later changed into three-year fellowships targeting beginning graduate students. In 1972, NSF began allowing students to spread that support over five years. And since 1988 recipients have had the option of deferring the fellowship for one year if they decide they need to take a breather from academia. The original purpose of the program, says Porter, was to support only the top students. "In the early years, we got the biggest bang for the buck," he says. But Sputnik jolted the nation out of its scientific complacency and, combined with President Johnson's subsequent desire to meet the needs of all students, the program became less elitist. At its peak, in 1968, NSF was supporting 2,500 fellows annually. But that outpouring of funds didn't last long. "Graduate student support all but dried up," says Porter about government-wide programs that at one point supported some 80,000 graduate students. By the early 1970s, a recession and changing attitudes about the role of the federal government led to the dismantling of all these programs except for the NSF fellowships. Even that program shriveled. With its budget slashed by 40 percent, NSF trimmed its support to 500 students a year as "an insurance policy that there would be a few good scientists," adds Porter. The outlook worsened in the early 1980s, when Congress beat back an attempt by President Reagan to cut out all money for new awards. Those dark days are now over, however. By the mid-1980s reports about an upcoming shortage of scientists began to appear, and NSF's fellowship budget began to swell again. And this time, the nation's changing demographics has forced NSF to assess how the program can better serve minorities and women. "The graduate fellowships were not reaching very many minorities," acknowledges Porter. "They were not very competitive." In 1978, NSF had begun setting aside 10 percent of its fellowships for U.S. citizens who are black, American Indian, Hispanic, Native Alaskan, or Native Pacific Islander. The agency awarded 43 of those fellowships that first year. Last year, NSF made it possible for fellows to go to school a few months early to get to know their advisers better. The lack of women is also being addressed. One obstacle, says Porter, is that too few women even apply. So this year NSF established a "Women in Engineering" program that adds 80 slots for female engineers. Within that number, a dozen slots are reserved for minority women. But more than the mix of students has changed. Today, there is considerably more variation in what the fellows have done or plan to do with their careers. In the past, students often went directly to graduate school after getting their bachelor's degrees. And more than two thirds of them then took positions in research universities. "But now they follow all sorts of nontraditional career paths," says Porter. Each spring, the National Research Council convenes a dozen panels - about 350 scientists in all - to evaluate applicants. Criteria include grade-point average, course work, Graduate Record Examination score, letters of recommendation, and research aspirations and experience. In 1989, the $34 million program provided about 2,000 recipients with $12,300 per year toward full-time graduate study, as well as an annual $6,000 to the institution to cover tuition and fees. Last year, 5,367 students applied for 760 open fellowships slots. That award rate has fluctuated over the past 15 years from 8 percent to almost 19 percent. A profile of last year's entrance class shows that 192 are studying biological sciences, 170 are engineers, and 144 are studying social sciences and psychology. In addition, 30 are mathematicians, 22 are studying applied mathematics, 50 computer science, 75 physics and astronomy, 54 chemistry, and 23 earth sciences. There are 271 women among those 760. Another 1,619 applicants received honorable mention, a sign that they would have been worthy of support had more funds been available. According to Porter, many graduate programs actively recruit these students as well. Also, 797 minority students applied, of whom 100 received minority fellowships and 251 received honorable mention. NSF supplements those awards with Mentoring Assistantships - three months' support during the summer so that the students can develop mentor relationships at their new institution. The minority program also gives $1,000 Incentives for Excellence Scholarships to the department from which the fellows come. That money is to be used as a prize for one or two minority undergraduates. Overall, NSF fellows still represent an elite group. In 1990, for example, NSF expects to support 17,000 graduate students; all but 2,000 will be assistants paid through grants, an arrangement in which they "do research under someone's tutelage and must perform services that are useful, but not sufficient to get a Ph.D.," says Porter. Fellows, on the other hand, "have this independent means of support. They do not have to work in areas where professors have research grants. These [fellowships] provide extra flexibility to the very best students." That flexibility is much appreciated. Harvard's Gould, who held an NSF fellowship while at Columbia University, says that the program turns money into precious time for scholars. Michigan's D'Arms, who has 55 fellows on his campus, notes that the fellowships "promote unimpeded progress and prevent a loss of momentum or spirit." NSF officials say they hope to have an even greater impact on future generations of scientists. They plan to double the number of new fellowships as Congress meets the president's goal of doubling the foundation's budget. For 1991, that means a request for $60 million and 1,100 new slots a year. The 10 percent set aside for minorities will increase to as much as 20 percent by 1991, along with growth in the program to support women graduate students in engineering. And once more, the program is becoming one of many to which graduate students can turn. About 56,000 of the nation's 300,000 graduate students are supported by the federal government. Presidential science adviser Allan Bromley plans to create a multiagency coordinating body to oversee the various programs, including NSF's. Some of the other fellowships offer more attractive packages than does NSF's. For example, the Department of Defense, the Office of Naval Research, and the Howard Hughes Medical Institute cover the full costs of tuition, while NSF promises just $6,000 a year to institutions. Other programs tried to provide a set tuition reimbursement, but universities balked at making up the difference, so those fellowships pay the full amounts. By contrast, "NSF, with its prestige, has come to be accepted at that level," says Porter. But NSF officials know that, even after 38 years, they must continue to spread the word. To commemorate its 25,000th fellow, NSF gave 18 societies $1,000 each to sponsor talks by former fellows. The list is impressive: So far, James Gibbons, dean of engineering at Stanford, addressed the American Society of Mechanical Engineers; Edward O. Wilson, Harvard's father of sociobiology, gave a presentation for the American Society of Zoologists; and Michael Posner of the University of Oregon will speak this summer at a meeting of the American Psychological Association. In February, a panel at the annual meeting of the American Association for the Advancement of Science discussed the future of the program. w For more information, contact: Terence L. Porter, Director, Division of Research Career Development, NSF, 1800 G St., N.W., Washington, D.C. 20550; (202) 357-7536.

NSF Fellowships Weather Politics, Promote Superb Science

Interested in reading more?

Become a Member of