To Compete or Collaborate? Career Questions in Science

May 25, 2001

Everyone knows the competition for limited research dollars is hot. And the cost of doing research, including the time it takes to develop proposals and grants, is on the rise. So where does that leave you, as a young scientist? If you're like most doctoral students, it leaves you under increased pressure to be more effective and efficient in your research.

Figuring out how to do that is tricky since it involves balancing competing interests: When should you cooperate with colleagues on your research and when should your competitive instincts take over? Should you pursue high-risk research or go for the almost-sure thing? Should you collaborate with others within your discipline or take the interdisciplinary route?

Let's look at each of these questions in turn and follow with a discussion of their implications for your scientific career.

Cooperative versus competitive science

As the competition for research dollars has intensified, some investigators have become less willing to share their scientific results with others. They also have become more critical in reviewing submissions to refereed journals, more disparaging of competing research proposals, and less willing to collaborate on research projects. Leon M. Lederman, the Nobel laureate and high-energy physicist, put it this way:

"Scientists are increasingly viewing their fellows as competitors, rather than colleagues, leading to an increasingly corrosive atmosphere. The manifestations of this attitude range from a reluctance to share new results with other scientists, to public bickering about relative priorities in funding different fields."

A biology professor at the University of Minnesota said it more directly: "I feel pressure not to share information about exciting data ... for fear that others, with larger labs, will be able to use this information to write a similar proposal."

Yet, the opposite behavior is also occurring. The development of new communication tools, such as the high-speed Internet 2, a national computer network available exclusively to researchers, is leading to greater collaboration, not only among members of the same department, but across departments and even universities.

Such tools have also made it possible for expensive facilities to be shared with users who are conducting their own experiments. The Kitt Peak National Observatory, near Tucson, is a good example. There, astronomers from all over the world, some of whom remain at their home institutions, share time on the many expensive telescopes at the observatory complex. Such cooperative ventures are occurring across science and engineering and are likely to become more common in the decade to come.

High-risk versus low-risk research

The heightened competition for research dollars has caused some scientists to shy away from high-risk research. These investigators play it safe by concentrating on less risky projects that they feel are more likely to succeed. By "high risk," I mean ideas for which there is little preliminary information to support a grant proposal -- for example, a venture into a new field or an attempt to develop a new model, a new organism, or a new instrument.

As a materials-science professor at Carnegie Mellon University noted, "We are tending to do 'safer' projects because we can't afford to have experiments not work."

However, taking the easy route can backfire. Ursula Goodenough, former president of the American Society for Cell Biology and a professor at Washington University in St. Louis, says that keeping risky ideas out of your grant proposals is not such a good strategy. "I have seen reviewers say, 'This is fine, everything ought to work, but the answers are boring,' and they give it a middle priority," she says. Better advice for young scientists, she says, is to propose a mix of some tried-and-true experiments and some neat and interesting ideas, with experiments that will address them.

Both the National Institutes of Health and the National Science Foundation have special programs to support what N.I.H. calls "high-risk/high-impact" proposals. These "sugar grants," in N.S.F. terminology, provide a small amount of money for a short period to allow an investigator to check out an interesting idea and develop it to the point where it could be submitted for a normal grant.

Disciplinary versus interdisciplinary collaborations

Today's scientists need to be both disciplinary and multidisciplinary. They need to have the breadth to see problems, and the depth to solve them. Richard Zare, a chemistry professor at Stanford University and former chairman of the National Science Board, calls this combination "T competency," where the horizontal bar of the "T" represents breadth and the vertical bar represents depth.

Mr. Zare makes it clear that multidisciplinary research can succeed only if it stands on a strong disciplinary foundation. Yet, more than ever, he feels that young scientists need a combined competence in selected disciplinary subfields and cross-disciplinary proficiency.

Industry, and some federal agencies, have responded to this need by supporting more interdisciplinary scholarship. "More and more federal funding agencies, such as N.S.F., the Department of Energy, as well as private industries, insist upon" interdisciplinary research, says Robert Shelton, vice chancellor of the University of California at Davis. "Many of them want a clustering of expertise at a single institution to solve problems."

What does all this mean for your scientific career?

First you need to understand that all of these tensions I've described are a normal part of the research enterprise. The key is to look for the balance appropriate to your specific circumstances and career stage.

It would be easy to say that you should simply do all of the above -- be competitive and collaborative, do high-risk and low-risk research, be disciplinary and interdisciplinary in your work. But not only is that impossible, it's not even desirable. Balance doesn't mean equality at all points in time. And at different stages in your career you are likely to be at different points on the continuum in each area.

Early in your career you will probably emphasize more competitive, low-risk research to establish your ability to be successful and deeply grounded in your discipline -- something that departments look for when granting tenure. Later, you can "afford" to be more collaborative, to take greater risks, and to contribute your established expertise to a multidisciplinary project.

Keep in mind that norms vary by discipline. High-energy physics, with its expensive equipment and large staff requirements, is inherently cooperative, so it is not unusual to find that you are just one of hundreds of collaborators on a single experiment. Some areas of biology are this way as well. A recent paper on the human genome in Nature had more than 1,000 co-authors. In such cases, finding ways to distinguish yourself from your collaborators becomes a key task.

In general, you need to assess the norms of your particular field and then look for ways to move beyond them, if only modestly at first. If you start out on the competitive/low-risk/disciplinary end of the scale, you can look for ways to test the waters with some relatively high-risk research in collaboration with more experienced colleagues, perhaps in a multidisciplinary context.

How do you find out about the norms in your field and where to place yourself on the spectrum? That's as easy as talking to colleagues, particularly senior scientists, at professional conferences and departmental seminars.

While finding your niche among all the various possibilities is a challenge, it can also be quite rewarding. If you regularly check to see where you are on the continuum and keep an eye out for opportunities on the other side of the seesaw, you will increase your likelihood of a varied and successful scientific career.

Richard M. Reis is director for academic partnerships at the Stanford University Learning Laboratory, and author of Tomorrow's Professor: Preparing for Academic Careers in Science and Engineering, available from IEEE Press or the booksellers below. He is also the moderator of the biweekly Tomorrow's Professor Listserve, which anyone can subscribe to by sending the message [subscribe tomorrows-professor] to

Have a question or a suggestion for Richard Reis? Please send comments to

To purchase Tomorrow's Professor on line, choose a bookseller:  Barnes & Noble