9b. Elements of a responsible publication

Each element of a publication serves an important purpose and must be carefully prepared to make sure it serves that purpose.

     Abstracts. Abstracts summarize the content of publications in sufficient detail to allow other researchers to assess relevance to their own work. Abstracts, therefore, should neither understate nor overstate the importance of findings. Negative results that might be important to other researchers or the public should be mentioned. The data presented in the abstract should be the same as the data presented in the body of the publication—an obvious requirement, but one that studies of publication practices show some authors do not follow (see Pitkin, Additional Reading).

To ensure completeness and accuracy, many journals now use structured abstracts. This assures that all of the key elements of the publication are mentioned and easily identified. With scientific publications now running in the millions per year in well over 100,000 journals, researchers cannot read all seemingly relevant publications in detail. They must rely on abstracts to point them to important developments and findings.

     Methods. Researchers cannot check and build on the work of others without knowing how it was conducted. Methods therefore should be described in sufficient detail to allow other researchers to replicate them. When researchers use well-established methods, this section of a publication can be shortened, provided appropriate references are given to a full description of the methods along with any changes that have been made. New or unique methods should be described in more detail to allow other researchers to replicate the work.

     Results. Research results should be reported in sufficient detail to allow other researchers to draw their own conclusions about the work. This does not mean that every piece of recorded data should be reported. Researchers can and must process their raw data before publication (to keep publications to a reasonable size if for no other reason). However, results should not be left out just because they do not agree with the conclusions the authors would like to reach. The results section should represent a complete summary of what was discovered, leaving interpretations for the closing discussion.

     Discussion. Researchers can and should evaluate the significance of their findings under discussion—also called conclusion or summary. This portion of a publication helps those who are less familiar with the field understand the importance of the findings. It also provides a venue for identifying unresolved problems and future research needs.

Since the discussion is read by individuals who may not be able to evaluate its validity, it is particularly important that authors avoid bias and one-sided reporting in this section. Cautions and other interpretations should be mentioned along with the limitations of the study to provide a balanced view of the reported results. Review articles (articles that survey research findings in particular areas) should make an honest effort to cover all relevant work. It is not always easy to recognize one’s own biases, which is a good reason to ask colleagues to read and comment on manuscripts before they are submitted for publication.

     Notes, bibliography, and acknowledgments. Notes, bibliography, and acknowledgments should be used to place publications in context and to give credit to others for their ideas, support, and work. They serve to:

• provide support for important statements of fact or assumptions,
• document the work of others used in the publication,
• point to additional reading and resources, and
• recognize the support of funding agencies or colleagues and staff who do not qualify as authors.

Since others rely on and trust this information, it, along with every other element of a responsible publication, should be fair and accurate.