Correcting the problems and inequities that bedevil postdocs will require "changing the culture" of labs and universities across the country. What's more, the National Science Foundation (NSF) can and should take steps to catalyze reform. This call to action emerged as the unanimous conclusion of an NSF workshop on the postdoctoral experience held 15 to 17 November at foundation headquarters in Arlington, Virginia.

Participants took care not to recommend specific actions or policies, but they agreed on seven "overarching themes" that they believe should guide the effort. Entitled "Postdoctoral Appointments: Policies and Practices," the workshop did not aim to "identify issues. ... We've spent 10 years identifying issues," Timothy Anderson, associate dean of research and graduate programs at the University of Florida, Gainesville, said in the charge to group. Rather, the gathering aimed to identify broad approaches and strategies that NSF can implement to broaden access to postdoc programs, facilitate young scientists' transitions into and out of postdoc positions, improve their career options and outcomes, and enhance their living standards and professional status.

Approximately 50 participants drawn nearly equally from faculty members, university administrators, postdocs, and representatives of industry, government agencies, and professional associations joined with key NSF staffers in plenary sessions and breakout groups. Within a schedule that gave all interest groups the chance to hear and react to each other's views, participants suggested, analyzed, and critiqued scores of ideas and possibilities as they gradually distilled their thinking down to seven broad concepts that won universal assent. Workshop members agreed that these ideas, if adopted by NSF, could spark needed change among mentors, funders, administrators, and postdocs.

At the foundation of the workshop's deliberations was a definition of a postdoctoral appointment: A postdoc, the workshop participants agreed, is, or at least ought to be, a temporary stage in a young scientist's progression toward a productive and sustainable career and not a stopgap job or a source of cheap scientific labor. The workshop participants also agreed that postdocs, mentors, and the wider research community must recognize explicitly that only a minority of postdocs can attain the traditional ideal career: a tenure-track faculty job at a research university. A number of careers that use scientific training--in industry, government, journalism, education, entrepreneurship, or other fields--must be recognized as successful outcomes for scientists. Mentors and universities, the participants emphasized, have the responsibility to help postdocs move toward these other satisfying professional futures.

The workshop asserted as its first theme the need for "professionally appropriate practices and compensation" in all postdoc appointments, including access to health coverage. In theme #2, it called for "modifying grant practices and policies" in order to draw principal investigators' and institutional officials' attention to the need to provide postdocs with a satisfactory training and employment experience. For example, NSF could seek information in grant applications, including renewals, about the services that the institution makes available to postdocs, the career outcomes of the lab's postdoc alumni, and the nature of the individual development plans to be created for the postdocs working on the project. An explicit plan detailing expectations and available resources is essential for every person in a postdoctoral position, the workshop participants agreed.

The third theme was a number of possible "new funding mechanisms," including, perhaps, more individual postdoc fellowships, as well as grants to foster postdoc offices on campuses or provide resources for training in what James Mitchell, a professor of chemical engineering at Howard University in Washington, D.C., termed "lifemanship skills" such as job hunting.

Theme four, "measuring outcomes," asserted the need to document systematically postdocs' experience in their mentors' labs and beyond. Tracking the length of time in their positions, their scientific productivity, and their subsequent career paths would create incentives for improvements in all these areas, workshop participants agreed.

"Disseminating information" (theme #5), could be expected to catalyze many improvements, the group noted. The idea of a unified Internet portal to centralize information and resources won widespread support. Whether run by NSF or some other entity, the portal could include a listing of all available NSF-supported postdoctoral positions; information on mentors, labs, and projects; links to networks of lab alumni; and other material. Open access to such information would help weaken the "old boy" hiring network, boost diversity in staffing, increase awareness of postdoc issues among mentors and administrators, and allow young scientists to make better choices about postdoc positions and career options.

Theme #6, "existing programs," calls on NSF to identify practices already in effect that can contribute to the goal of change. The final theme, "institutional policies" (#7), encourages the institutions in which postdocs work to establish and publish clear policies on employment issues including pay, benefits, length of appointment, career development resources, and working conditions. Finally, the workshop called for "sustained marketing" to win broad attention and acceptance for any reform initiatives NSF undertakes.

The appropriateness, feasibility, and possible favorable and unfavorable consequences of each proposal engaged the participants' sustained attention. They recognized that many suggested changes would cost money and impose bureaucratic burdens. In addition, they acknowledged the reality of tight resources. For that reason, many agreed with Alyson Reed, executive director of the National Postdoctoral Association, that encouraging establishment of postdoc offices on campuses was the proposal that could have the "broadest possible impact" for the dollars spent. Such offices "highly leverage" investment, Anderson noted. They "create an advocate for postdocs on campus" in addition to coordinating services and information, added Philip Clifford, associate dean of the Medical College of Wisconsin in Milwaukee.

Collecting and disseminating information about postdocs' experiences in various labs and in their later careers are other highly cost-effective ways to encourage change, the group agreed. Doing so could also help "build best practices of industry" into academe and NSF, said Geoffrey Davis, visiting scholar at the international scientific research society, Sigma Xi. NSF already has "similar kinds of [reporting] requirements for live animals," Davis added.

A formal report on the workshop will appear during the winter. But no document can in itself bring change, warned meeting organizer Robert Lichter of Merrimack Consultants in Atlanta, Georgia. "We can all sit here and say this is wonderful," but others in the scientific community may not buy in, he said. A similar meeting about postdocs held at the National Institutes of Health (NIH) in October 2003, has yet to produce any results, and its "momentum may have died," said Clifford, who added that a "blockbuster proposal" from NSF could revive NIH's effort. The workshop report can be a "vehicle" of reform, Lichter concluded, but change will only come from many "individuals saying this is important."

Beryl Lieff Benderly writes from Washington, D.C.