Scientific Writing: Peer Review and Scientific Journals

by Natalie Kuldell

Visionlearning has updated this module since its original publication. For more comprehensive information on scientific writing, please go to Scientific Communication: Understanding Scientific Journals and Articles.
Figure 1: Magazines.

We’ve all read the headlines at the supermarket checkout line. “Aliens Abduct New Jersey School Teacher” or “Quadruplets Born to 99-Year-Old Woman: Exclusive Photos Inside.” Journals like the National Enquirer sell copies by publishing sensational headlines, and most readers believe only a fraction of what is printed. A person more interested in news than gossip could buy a publication like Time, Newsweek or Discover. These magazines publish information on current news and events, including recent scientific advances. But where do they get their scientific information? News magazines hire people to read specialized journals where scientists publish their research findings. Scientific journals can provide reliable information because of the process called “peer review,” in which other scientists (peers) evaluate the value and credibility of research before allowing it to appear in print.

Peer Review

Peer-review is performed by knowledgeable scientists who are not directly involved with the research being evaluated. In fact, reviewers are often scientific competitors! To remove any bias from the review process, most manuscripts (articles prior to publication) are independently considered by three reviewers. Reviewers consider the validity of the approach, the significance and originality of the finding, its interest and timeliness to the scientific community, and the clarity of the writing. Reviewers then provide feedback on the manuscript they have read. Journal editors rely on peer-review feedback to guide their publication decisions, and authors use reviewers' comments to refine the text of their manuscript and the experiments within. Journal editors must occasionally resolve issues related to conflict of interest among reviewers; reviewers’ identities are generally not revealed to manuscript authors. This later rule is intended to free reviewers from any social pressures, allowing them to consider only the quality of the science before them.

Reviewers are expected to keep the information in a manuscript confidential until it is published, but it is rare that the work comes as a complete surprise to the entire scientific community. This is because peer review is integrated into almost every step of science. Most research scientists request public funding for their experiments. Funding decisions are made by a committee of other scientists who debate each proposal’s likelihood of success, the validity of its approach, and the importance of the question being asked. Once funded, the experiments can begin, and preliminary data is often revealed at scientific meetings. This allows the findings to be debated and defended with colleagues prior to publication. Once the experiments are completed, a manuscript is written and circulated to all those who contributed to the work. Manuscripts commonly undergo several rounds of revision by the authors before being submitted to a journal for peer review. Journals vary in their selectivity and focus. For example, the journal Cell publishes articles focused on biological process related to cell function; the journal Environmental Science and Technology publishes articles related to environmental pollution and cleanup. Consequently, manuscripts are first sent to the most widely read journal likely to publish the work. If that journal declines to publish the manuscript, it can be sent to a different journal for consideration.

Figure 2: Reconstruction of the Piltdown man skull.

Thanks to the peer-review system, readers can feel confident that the information found in scientific journals is credible. Despite the best efforts of reviewers, cases of scientific misconduct do occur and incorrect or unsubstantiated data does get published. Some cases turn out to be elaborate hoaxes. For example, in 1912 Charles Dawson showed off parts of a skull and jawbone to the public and convinced scientists that the fossils represented the missing link between man and ape. Dawson’s “Piltdown Man” confused the scientific community for 40 years until it was discovered that the skull was only 500 years old rather than 500,000, and the jawbone was that of an orangutan. In other instances of misconduct, data in scientific journals has been improperly reported. Twenty-five papers published by Jan Hendrick Schon between 2000 and 2003 were called into question because data from key experiments related to superconductivity were inadequately documented and several control samples were improperly reported. At least sixteen of those papers have since been declared to be false, and the journal Science has withdrawn eight of Schon’s papers. The scientific community has also been tarnished with cases of outright fraud and forgery. For example, in 1974 William T. Summerlin used a felt-tip pen to change the fur color of some mice in an effort to convince his mentor that he had successfully grafted skin between different strains of mice.

Cases of scientific misconduct are rare but important because of the publicity they receive once they are discovered, eroding the public's trust in the peer-review system and science itself. One cornerstone of science is that scientific findings must be reproducible and well documented. Many instances of scientific misconduct have been exposed when other scientists cannot reproduce reported data. For example, in 1989 Stanley Pons and Martin Fleischmann announced to the world that they had discovered “cold fusion,” a possible unlimited source of energy. When other scientists were unable to replicate the work, it was discredited. To keep this type of conduct in check, scientific articles include detailed descriptions of experimental protocols that enable others to reproduce experiments.

Article format

Scientific articles are neither mystery stories, intended to surprise the reader with a discovery, nor diaries of every successful and unsuccessful experiment performed. Instead, the major discoveries are set out in the article’s first section. The writer’s goal is to present his or her findings and persuade the readers with his or her interpretation of the data. Journal articles are divided into sections, each with a specific purpose. Although every journal has a particular “house style” for the way it formats and names the sections of the scientific articles it publishes, a generic scientific article would include the following:

  • Title: A well-written title gives information about the research. The names of all scientific contributors are usually listed immediately after the title. By convention, the scientist who performed most of the work described in the article is listed first, while the last author is usually a senior scientist who secured funding for the work and who runs the lab in which it was performed.
  • Abstract: Some journals call this the “summary” because it must concisely describe the experimental question, the general methods and the major findings and implications of the experiments. It is generally limited to 250 words because the text is typed into literature databases, such as PubMed (see the Research links in the right menu). A well-written abstract will include likely “search words” because searching databases is a major way scientists find articles of interest.
  • Introduction: The central experimental question and important background information are presented in this section. Relevant and established scientific knowledge is cited in this section and then listed in the References section at the end of the article. Introductions are intended to lead the reader to understand the authors' hypothesis and means of testing it.
  • Materials and Methods: All experimental procedures and reagents are described in detail sufficient for another researcher to reproduce the findings. This section must be accurate and complete if the discoveries are to be validated and then extended by others.
  • Results: The data are presented in this section, giving other scientists an opportunity to judge their merit. The findings are described with words and also illustrated using figures and tables. Figures are used to facilitate the interpretation of the data and have accompanying explanations, called “legends.”
  • Discussion: In this section, authors may present a model or idea they feel best fits their data. They also present the strengths and significance of their data. Some journal articles fuse “Results and Discussion” into one section, but when they are separated a reader can easily distinguish the data collected from the authors' interpretation of it.
  • References: Scientific progress builds on existing knowledge, and previous findings are recognized by directly citing them in any new work. The citations are collected in one list, commonly called “References,” although the precise format for each reference section varies considerably. Some journals ask that citations be listed alphabetically, whereas others require that they be listed by the order of appearance in the text.

Figure 3: Scientific research articles often include figures and tables that help explain the results of the experiments conducted during the work. This figure was modified from P. M. Vitousek, and P. A. Matson (1993). "Agriculture, the Global Nitrogen Cycle, and Trace Gas Flux." The Biogeochemistry of Global Change: Radiative Trace Gases, R. S. Oremland, New York, Chapman and Hall: 193-208.

Reading scientific articles

The specialized format of scientific articles can make them hard to approach, but once the purpose and requirements for each section are understood, primary journal articles can be read by anyone. Strange as it seems, the most efficient way to tackle a new article may be a piecemeal approach, reading some but not all the sections and not necessarily in their order of appearance. For example, the abstract of an article will summarize its key points, but this section can often be dense and difficult to understand. Sometimes the end of the article may be a better place to start reading. In many cases, authors present a model that fits their data as the final figure of the article. The discussion section may emphasize some themes or ideas that tie the story together, giving the reader some foundation for reading the article from the beginning.

Readers who are unfamiliar with a specific field may feel like outsiders when they try to read scientific articles. Sometimes older research articles, from the time when a field of study was first developing, can be useful because they present the concept in a simpler fashion (look at the Classics section in the right menu for ideas). Written guidelines for reading can also help if they are written by an experienced reader. The reading guidelines can demystify the process of reading an article. For example, focus on the anatomy of the paper (e.g. “How many references are cited, and how are they organized?”) and the language within (e.g. “Use your glossary to find definitions for the following terms...”). Guidelines for beginning readers can be content oriented, highlighting interesting scientific questions and approaches, or even the quality of writing.

Figure 4: Cover of Nature magazine.

The scholarly vocabulary in scientific articles can be one of the greatest obstacles to reading the primary literature. In some articles it’s possible to pinpoint the exact moment the authors move from understandable language to “science speak.” A list of terms and definitions can help a new reader navigate this transition. Another helpful tool is a well-written and relevant review article. Particularly good review articles can be found in the Journal of the American Medical Association, Science and Nature, and other journals. Review articles can help a reader take his or her first steps into the primary scientific literature.

In addition to the language barrier, scientific articles often contain daunting figures or formulas. A “jigsaw” approach can be an effective way to understand these figures if the reading is done as a class. Different readers work individually or in groups to become experts on just one figure of an article. To reassemble the article, each reader or group presents their figure, allowing everyone to benefit from their expertise but alleviating the need for everyone to master all aspects of the reading. For more information on how to read and understand figures, see our Visualizing Scientific Data module.

Given the volume of literature that exists and the number of new research articles available each day, finding appropriate articles should be possible. Reading them is a valuable and lasting skill. It emphasizes science as a process of discovery. It also prepares anyone interested in science for life-long self-directed learning. Scientific articles can also teach persuasive writing, a valuable tool for any field of study. If you would like to learn more about scientific research papers, read "The Case of the Ivory-Billed Woodpecker" linked in the Research menu.

Further Readingtoggle-menu

Natalie Kuldell “Scientific Writing” Visionlearning Vol. SCI (2), 2004.

Don't ask me who's influenced me. A lion is made up of the lambs he's digested, and I've been reading all my life.
-Giorgos Seferis, 1900-1971

Want ad-free content? Registration is free »