NSF Stresses Publication Quality, Education With New Grant Format

WASHINGTON—The National Science Foundation has changed its format for proposals in a strong message to applicants about what their priorities should be. The new format reflects the nation’s increasing concern about the training of new scientists and information overload. On October 1, NSF began requiring that all proposals include a statement about a proposed project’s educational potential and a list of the young scientists being trained in the applicant’s laboratory. In addition, NSF declared that only 10 publications by then principal investigator would be considered during the merit review. “Scientists themselves have to recognize that science is getting out of hand,” says Margaret Davis, Regents Professor of Ecology at the University of Minnesota, about the effects of increased specialization and sprawling laboratory teams on the future of research. Too often, she says, lab chiefs have too little time to work with graduate students and no time at all to help undergraduates. A good teaching record at most universities is worth less in recognition and career advancement than a million-dollar grant, says Mary Clutter, who heads the NSF directorate of biologic, behavioral, and social sciences. And one way to succeed at the grantsmanship game is to hire technicians instead of sponsoring graduate students. “I think the federal government has colluded with that,” adds Clutter, who shepherded the changes at NSF. “It’s time to get back to basics.” The new NSF policy-requires applicants to specify, in a separate statement, how the research proposed will help train undergraduates, graduate students, and postdoctoral fellows. They are asked to give particular attention to minorities and women. NSF also Wants to know how the proposed work will affect course preparation and seminars. In addition, principal investigators must name the graduate students and postdoctoral fellows they supported during the past five years, and provide a summary of the total number of young scientists supported and advised. “This sends a signal to the community that we value education,” says Clutter. The new information will be incorporated into one of the four criteria that NSF uses to select from among the more than 20,000 proposals it receives each year. That category, which measures the work’s contribution to the scientific infrastructure, is one way that NSF separates proposals that already excel on scientific grounds. With so many proposals earning high scientific ratings, say administrators, this statement could be a deciding factor in whether to award funding. But scientists are not so sure. “I don’t think it will matter at all,” says Linda B. Smith, professor of psychology at Indiana University, Bloomington, and an NSF reviewer; She expects the effect to be more subtle—and more gradual. “If a lot of the graduate students don’t stay in the field or don’t go on to good labs, that will color my views of the labo ratory and the grantee,” she says. The changes are also meant to address the overwhelming pressure to “publish or perish.” “People are writing too many papers of perhaps lesserquality,” says Clutter about the phenomenon, which some scientists believe can contribute to shoddy or even fraudulent research. The new NSF guidelines require that biographical sketches list up to five publications relevant to the proposed research plus five other publications, rather than listing all publications for the past five years. Since this much abbreviated curriculum vitae makes it difficult for NSF to know who might be biased in favor of the applicant, investigators must also list long-term collaborators and coauthors with whom they have worked in the past 48 months, plus their graduate and postdoctoral advisers. The idea of selecting five papers puts a greater burden on the scientist to decide what is important. The curriculum vitae of most investigators is so thick with publications, says Smith, that “you can’t find the real stuff from the irrelevant stuff.” NSF officials and researchers alike hope this requirement, part of a trend exemplified by Harvard Medical School’s decision last year to consider just five articles in tenure decisions, will yield longer, more comprehensive articles and reduce the number of repetitive papers—or least-publishable units, as the articles are sometimes referred to. The National Institutes of Health has had a similar limit in effect for about four years: Extramural grantees seeking to renew their grants are allowed to include no more than 10 manuscripts as appendices to their applications. Although NSF reports positive responses by scientists to the changes, it will take a long time to know what the real impact will be. “It will be Pollyannaish to think that scientists will welcome these changes at first,” says Smith. Any change can be frustrating to a scientist already spending too much time on a grant proposal, she notes. “It’s not something that research scientists have focused on,” says Bernard McDonald, NSF deputy division director of mathematical sciences, about the educational statement. He has even fielded calls from scientists who wondered how their abstract mathematical endeavors could be called “educational.” Some scientists worry that the new policy may penalize them for failing to recruit and retain graduate students. Others may resent any limit on publications. “For people who have hundreds of papers, I suppose they will gnash their teeth,” says Davis. “But for those people who don’t have a long bibliography, it comes as a relief.” At the same time, one NSF reviewer and grantee notes that most reviewers ask someone whose work is closest to that of the grantee for their opinion rather than sorting through dozens of articles. Clutter expects it will take a while for these changes to sink in and have their desired effects. The directorate for mathematical- and physical sciences, which receives many of its proposals during early fall, decided not to make the changes mandatory until after January 1990 so it would have more time to publicize them. On the other hand, applicants in the biological, behavioral, and social sciences who fail to follow this new format will be advised to update and resubmit their application. “Every proposal that BBS reviews will have these changes,” promises Clutter.

NSF Stresses Publication Quality, Education With New Grant Format

Interested in reading more?

Become a Member of