It's a big day when you have enough results to publish a paper. Not only is it great to see your name in print, but a published paper is also tangible evidence of your hard work and an important step in your career. But first you have to sit down and write the thing.

Many people are great in the lab but drag their feet when it comes to writing up their results. But even if someone else--your adviser, say, or another senior scientist--offers to do the writing for you, don't take them up on their very tempting offer. Insist they let you give the writing a try. You need all the writing practice you can get, and it will help you further down the road when it's time to tackle your 100-plus page thesis.

Why write?

Keep in mind that the purpose of your article is to provide other researchers in your field with specific information:

  • The question(s) you asked at the outset.

  • The experiments you performed to answer those questions.

  • The type of data you collected and your methods.

  • The conclusions you drew from your data and your suggestions for further research.

Your writing should reflect the same clarity and concision as your research. Don't rely on obscure language and jargon. Part of being a good scientist is being able to write up your results in clear and simple terms.

A well-written scientific article will address the statements listed above. The standard format found in nearly all peer-reviewed papers will help you organize your material into a logical order. The traditional format for a scientific paper has the following components:

  • Abstract

  • Introduction

  • Materials and Methods

  • Results

  • Discussion/Conclusion

These days, though, this format can vary widely, depending on the type of research article you're writing--a research letter versus a full-length research article, for example. Sometimes all of these basic components need to be included, just in a different order, and sometimes without the explicit subheadings. There are also small variations from field to field and journal to journal, so be sure to consult the information for authors for your target journal and make note of any specific requirements for organizing and formatting your text.

Title

Writing a good title is an art in itself. The title should give the reader a clear idea of what your experiment was about, i.e., the “what” and “how” of your work. Give your article a strong title for maximum impact.

Abstract

Write this section last and revise it meticulously. It may be the only part of your paper that gets read, so it must contain a precise (and complete) summary of your work.

The abstract is a one-paragraph summary, usually less than about 200 words, of the work described in the article. It should contain the following elements:

  • The central question (purpose) of the study.

  • A brief statement of what was done (methods).

  • A brief summary of the results.

  • A brief statement of the conclusions.

Many computer search engines make use of the information in the abstract, so make sure the key words are included so that an Internet search will pick up your article.

Introduction

Many people find this section the most difficult part of the paper to write. The goal of the introduction is to put your work in context. You need to include enough background information so that a reader not familiar with the field can understand the relevance of your work and how it fits into the big picture. And it has to pack this information into a small space. This means crisp, concise sentences.

The Introduction is where you state the purpose of your work. What research question were you attempting to answer? Why did you choose this experiment? What's the point?

You also need to mention previous research on or related to the question you're asking. How did what is already known help you in planning your own experiments?

Finally, be sure to state clearly your hypothesis or research question and your objectives. Read the introductions of several well-written papers in your field to get an idea of content and style. Some journals allow you to put the paper's main conclusion at the end of the introduction. Make use of this convention when you can, as it will prepare the reader for the main body of your article.

Materials and methods

In this section, accuracy is key, and details count. What was your experimental setup? Which type and brand of equipment did you use? How and when was the equipment calibrated? Which chemicals did you use? (Even the company you ordered them from and the batch number can be important.)

Be sure to provide enough information so that other researchers can assess the validity of your work and perhaps replicate it if they choose. Again, study several well-written articles from respected journals to get a sense of what to include . If you already know which journal you plant to submit to, study the articles in a recent issue to get a feeling for content and style. Common practice is to describe methods using the passive voice: "The sample was heated to 50 oC" rather than "We heated the sample to 50 oC."

Results

In the results section, you present the analyses of the data you collected. Graphs and tables are the easiest way to present this information. If your experiment resulted in a lot of numbers, try various types of graphs and tables to see which is the most effective in conveying your results.

A combination of tables and graphs usually works best, so that the reader can see both actual numbers and a graphical presentation of the relationship between variables. The Results section must closely match the Materials and Methods section. For example, if you present temperature data in the Results section, then the Materials and Methods section should say when and how you measured temperature.

Don't confuse the Results section with the Discussion section. The Results section should contain only the data you measured, as well as the graphic and tabular representation of the relationships between variables. Wait for the Discussion section to mention the patterns or trends you discovered. Creative data presentation can also highlight the trends you wish to further expound on in the Discussion section.

Discussion

Here's where you’ll put your analysis and interpretation of the results. In a nutshell, the Discussion section explains the meaning of the results. For example: If temperature affected the fertilization of zebra fish eggs, why did it do so?

You are allowed a little leeway here, but don't get too carried away with assumptions and wishful thinking. Be prepared to back up your analysis with solid evidence as presented in the Results section or by citing previously published work. Also, be careful that you don't include in your analysis a piece of data that you neglected to mention in Results. Journal editors and peer-reviewers are trained to look for such errors.

In addition, the Discussion section provides space for you to answer the questions you posed in the introduction. Did you discover what you thought you would? Were the results different from what you expected? What have you learned from your analysis? How does your work relate to other work in the field? Does it confirm or refute existing ideas and theories? What kinds of conclusions can you draw from the results? Ideally, the paper should illuminate and add to the development of what is known in a particular field

The Discussion section is also the place for suggesting ideas for future research. You may have answered some of the questions you started out asking, but most certainly the work you carried out has led to new questions. Pose those new questions here. It will provide possible new leads for other researchers as well as for you.

References

A list of cited literature, usually under the heading "References," is the last section of the paper. Be sure to follow the appropriate reference citation style for your target journal, as described in its Instructions to Authors. It is important to appropriately cite another researcher's work in the text whenever you refer to his or her results, conclusions, or methods. It is extremely important to mention the author when citing other’s work, so that it will be clear that a statement refers to someone else’s work and not your own.

Revising the first draft

Voila! You're finished. It's time to celebrate, right? Not quite yet. Once you have written the first draft of your article in the format described here, it's time to take a well-deserved break.

Print out your article and put it aside for a few days so you can get some much-needed distance from the process. In the next stage, you will have to switch hats: Put away your writer's hat and put on your editor's cap.

When you're ready to return to it in a few days, read it all the way through with a cold and critical eye, just like a reviewer and eventual reader will do. Keep an eye out for poor writing as well as holes and sloppy thinking. Don't be lazy about this step as any sloppiness on your part will be spotted by your peer-reviewers, and the paper will be sent back to you for corrections and alterations--or outright rejection. An extra round of edits may take you an hour or two, but another round of reviewing can take several months.

It isn't easy at first, but the best way to write a good scientific article is to do it by yourself. Once you've written a first draft, feedback from your co-authors, and your adviser, is essential. At this point, it's no longer just your paper. Actually, it never was--but someone has to write the first draft. All the other authors whose names appear on it must agree with the contents--and getting that to happen is a major challenge. But if you've done your job well by writing a stellar first draft, there should be little room for disagreement on the paper's essential elements.

Reaching a consensus on the details can require many drafts, and then the journal's review and revision process may add weeks or months to the process. But writing your first paper is an important accomplishment in your career as a scientist, so pat yourself on the back for having the courage to tackle this challenging task.

Patricia Gosling and Bart Noordam are the authors of Mastering Your PhD: Survival and Success in the Doctoral Years and Beyond (Springer, 2006). Dr. Gosling is a senior medical writer at CMP Medica, Malaysia, and also works as a freelance science writer. Dr. Noordam is professor of physics at the University of Amsterdam, the Netherlands, and director of a Regional Audit Organization. He has also worked for McKinsey & Co.

Comments, suggestions? Please send your feedback to our editor.

DOI: 10.1126/science.caredit.a0700051

Patricia Gosling and Bart Noordam are the authors of Mastering Your Ph.D.: Survival and Success in the Doctoral Years and Beyond (Springer, 2006). Gosling is a senior medical writer at CMPMedica in Malaysia and a freelance science writer.

Bart Noordam is a professor of physics at the University of Amsterdam, the Netherlands, and director of a regional audit organization. He has also worked for McKinsey & Co.

10.1126/science.caredit.a0700051