Getting Published as a Graduate Student in the Sciences

November 24, 2000

To succeed you must make your talents well known and widely appreciated. Publishing provides you with an important way to accomplish that. Your papers, available in libraries around the world, represent not only your product but also your résumé. -- Peter J, Feibelman, author of A Ph.D, is Not Enough! A Guide to Survival in Science

All professional scientists, whether in academe, industry, or the government sector, are expected to publish the results of their work in one form or another. But in order to land a job in any of these realms, you have to begin amassing a publishing record while you're still in graduate school.

Success in publishing requires the right attitude. You need to see publishing as a natural outcome of your dissertation scholarship -- a way to take some of the credit for what you have done and to communicate the results of your hard work to your future colleagues. Treat publishing as a forethought, not an afterthought.

What is a reasonable number of publications for a graduate student? First, keep in mind that in general the number of scientific publications can vary quite a bit by field depending on the nature of the problems under investigation. For example, some experimental fields such as microbiology and organic chemistry lend themselves to shorter investigations that can be reported more frequently, whereas other areas like population biology and high-energy physics, may take longer to produce publishable results.

There is also the question of how far you should go in "chopping up" your research into the smallest possible publishable units (jokingly called a "publion" in physics) so as to increase the total number of such publications. The editors of the journals to which you submit your papers will be the ultimate judges in this case. Mr. Feibelman takes the view that -- where possible -- shorter, more frequent papers that can later be combined into a significant review paper is a good approach for beginning researchers.

Most graduate students who publish do so as co-authors on articles with more senior researchers and faculty members. This is fine. Be open to sharing credit with others, even if you feel you have made a disproportionate contribution to the research. Contributions can be made in different ways and at different levels.

With all of the above in mind, Ph.D. advisers in most science and engineering fields recommend that a graduate student have at least one paper in which the student is named as the first author. Many professors also like to see their students as co-authors of one other paper.

What does first authorship mean? Usually it is the person who conceived, designed, and carried out the experiments and then took the lead in writing up the results. Since Ph.D. dissertations, by definition, represent an original contribution to the field by the student, it stands that most papers resulting from such dissertations should have the graduate student as the first author.

However, different norms do exist within different disciplines with respect to whose names appear as authors and in what order. In some cases, the lab director's name goes on every paper, says Donald Kennedy, editor of Science. "In genetics and microbiology, for example, things tend to be 'shared' and the director of the laboratory is almost always on the list of authors even if he or she did no direct work on the project," he says. "In population biology it is the people who actually do the work, usually graduate students, who are the only names on the paper."

It is important, therefore, for faculty advisers to make clear from the start what they expect of their graduate students and postdocs in the laboratory.

While you want to be sure to receive credit as a co-author for your contribution, be careful not to go over the line in this regard. Some advisers have been known to give students "complimentary authorships," by putting a student's name on a paper as a career boost even if he or she did little or no work on the research. One important test of co-authorship is whether or not everyone listed can give a talk, and answer subsequent questions, on the paper at a professional conference or symposium. Be certain you have that ability before accepting credit as a co-author.

Indeed, it is worth keeping in mind that the other side of credit is blame. If the paper is of a poor quality, or if the data are unreliable, as a co-author you will share some of the responsibility for that. If you have any concerns in this regard, you should discuss the situation with the senior author of the paper before it is submitted to a journal. While publishing is, as Mr. Feibelman points out, "a timeless advertisement for yourself," remember that you can't take back a publication. Excellent papers will serve as a permanent public record of your accomplishments, poor papers will damage your reputation. I'll touch on what makes for a poor paper below. Your goal must be to publish the highest quality research papers.

Most scientific publishing falls into one of three categories:

  1. peer-reviewed papers in journals,
  2. conference papers, and
  3. research reports.

While there are exceptions, peer-reviewed papers are generally of a higher quality than are those that appear in other forums. These are the papers that count most when reviewing a publication record for hiring, promotion and tenure. Conference papers, which cover in more detail the topic you are going to present orally at a conference, may or may not be peer-reviewed. Research reports are those usually required by funding agencies or your employer. Such reports often are more detailed than the journal or conference papers, but generally have a more limited audience. Of course they are often the basis for the first two types of papers.

So what is involved in actually writing publishable papers? We can only touch on a few of the key elements here. Fortunately, the graduate divisions of most research universities have pamphlets, guidebooks, and even workshops that can provide you with additional guidance. Two particularly good sources are The New Professor's Handbook: A Guide to Teaching and Research in Engineering and Science (Anker Publishing, 1994), and Graduate Research: A Guide for Students in the Sciences (Plenum Publishing, 1990).

Many experienced authors suggest that you organize a research paper somewhat as you would a newspaper article -- that is, tell the same story several times by going into increasing levels of depth and difficulty. Start by paying considerable attention to the title since it will determine if busy readers will go further. For example, "Sending Signals: How Bacteria and Plants Talk to Each Other," has a lot more punch than "The Mechanism of NodD1-Mediated Transcription at Nod Gene Promoters." The title needs to be concise, accurate, and compelling.

Next comes the abstract, which should be 50 to 300 words. It is often circulated much more widely than the article itself, so pay close attention to it. Cliff Davidson and Susan Ambrose, the authors of The New Professor's Handbook, refer to two types of abstracts: descriptive, which list the contents of the paper, and informative, which describe the most important research results and their significance. Most peer-reviewed papers should use the latter type.

The abstract is followed by the introduction. According to The New Professor's Handbook, the introduction serves several purposes: It describes the general topic area of the paper, lists the specific problems of interest, and presents the motivation for the work. Unless the manuscript is very short, the introduction should also include statements about the organization of the paper: Listing the major sections helps the reader understand the flow of ideas that follow.

Next comes the literature review where citing the work of others is essential. Obviously you want to show the connection between their work and yours. Not only does your professional integrity demand this, to do otherwise would be a case of fraud. It also certainly doesn't hurt to give credit to future colleagues, some of whom will be reviewing your paper for publication.

Now you need to discuss your research objectives and your methods of achieving them. Describe the equipment and experimental procedures you used for laboratory or field work, and the mathematical relations and solution techniques for a theoretical study. Then in the results section, which usually follows, you should refrain in most cases from including raw data but rather present the results themselves with some explanation of how you reached them.

Next comes the discussion section, where you explain the significance of your results. According to Mr. Davidson and Ms. Ambrose, a poor job here is the main reason for rejection of most journal papers. As they put it:

In some cases, there is a fatal flaw: the research results are simply not significant enough to warrant publication. In other cases, the findings are interesting and worth publishing, but the discussion is inappropriate. For example, the author may be afraid to make a bold statement even when it is supported by the data (perhaps the true significance of the work is not recognized), or conversely the author may make unsubstantiated, sweeping claims when in fact only modest claims are warranted.

Some papers follow the discussion section with a section on future work. If this is the case, such statements should be limited to broad overviews of the directions you see your research taking, and not the kind of detail that would fit into a forthcoming proposal.

Next to the title and abstract, the section most likely to be read by most readers is the one with the conclusion or summary statement. A conclusion states the outcome of your work whereas a summary is a brief statement covering the main points of the paper. Either, or both, may be found in a research paper.

While there are many references on writing research papers, don't forget what may be your most important resource, your research advisers. Asking advisers and other colleagues to give you critical feedback on your drafts before submission to a journal can save you tremendous time and will go a long way toward increasing your chances of eventually having your paper accepted for publication.

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