What Is and how organize a Scientific Paper?

1. DEFINITION OF A SCIENTIFIC PAPER

A scientific paper is a written and published report describing original research results. That short definition must be qualified, however, by noting that a sci­entific paper must be written in a certain way, as defined by tradition, editorial practice, scientific ethics, and the interplay of printing and publishing pro­cedures.

To properly define “scientific paper,” we must define the mechanism that creates a scientific paper, namely, valid (that is, primary) publication. Abstracts, theses, conference reports, and many other types of literature are published, but such publications do not normally meet the test of valid publication. Further, even if a scientific paper meets all the other tests, it is not validly published if it is published in the wrong place. That is, a relatively poor research report, but one that meets the tests, is validly published if accepted and published in the right place (a primary journal or other primary publication); a superbly pre­pared research report is not validly published if published in the wrong place. Most of the government literature and conference literature, as well as institu­tional bulletins and other ephemeral publications, do not qualify as primary literature.

Many people have struggled with the definition of primary publication (valid publication), from which is derived the definition of a scientific paper. The Council of Biology Editors (CBE), now the Council of Science Editors (CSE), arrived at the following definition (Council of Biology Editors 1968, p. 2):

An acceptable primary scientific publication must be the first disclosure containing sufficient information to enable peers (1) to assess observa­tions, (2) to repeat experiments, and (3) to evaluate intellectual processes; moreover, it must be susceptible to sensory perception, essentially per­manent, available to the scientific community without restriction, and available for regular screening by one or more of the major recog­nized secondary services (e.g., currently, Biological Abstracts, Chemical Abstracts, Index Medicus, Excerpta Medica, Bibliography of Agricul­ture, etc., in the United States and similar services in other countries).

At first reading, this definition may seem excessively complex, or at least verbose. But those who had a hand in drafting it weighed each word carefully and doubted that an acceptable definition could be provided in appreciably fewer words. Because it is important that students, authors, editors, and all others concerned understand what a scientific paper is and what it is not, it may be helpful to work through this definition to see what it really means.

“An acceptable primary scientific publication” must be “the first disclosure.” Certainly, first disclosure of new research data often takes place via oral pre­sentation at a scientific meeting. But the thrust of the CBE statement is that disclosure is more than disgorgement by the author; effective first disclosure is accomplished only when the disclosure takes a form that allows the peers of the author (either now or in the future) to fully comprehend and use that which is disclosed.

Thus, sufficient information must be presented so that potential users of the data can (1) assess observations, (2) repeat experiments, and (3) evaluate intel­lectual processes. (Are the author’s conclusions justified by the data?) Then, the disclosure must be “susceptible to sensory perception.” This may seem an awk­ward phrase, because in normal practice it simply means published; how­ever, this definition provides for disclosure not just in terms of printed visual materials (printed journals and the no longer widely used media called micro­film and microfiche) but also in nonprint, nonvisual forms. For example, “pub­lication” in the form of audio recordings, if that publication met the other tests provided in the definition, would constitute effective publication. And, certainly, electronic journals meet the definition of valid publication. What about mate­rial posted on a website? Views have varied and can depend on the nature of the material posted. For the most current information, consult materials from professional organizations and journals in your field.

Regardless of the form of publication, that form must be essentially perma­nent (often not the case for websites), must be made available to the scientific community without restriction (for example, in a journal that is openly accessi­ble online or to which subscriptions are available), and must be made available to information-retrieval services (Biological Abstracts, Chemical Abstracts, MEDLINE, etc.). Thus, publications such as newsletters, corporate publications, and controlled-circulation journals, many of which are of value for their news or other features, generally cannot serve as repositories for scientific knowledge.

To restate the CBE definition in simpler but not more accurate terms, primary publication is (1) the first publication of original research results, (2) in a form whereby peers of the author can repeat the experiments and test the conclu­sions, and (3) in a journal or other source document readily available within the scientific community. To understand this definition, however, we must add an important caveat. The part of the definition that refers to “peers of the author” is accepted as meaning prepublication peer review. Thus, by definition, scien­tific papers are published in peer-reviewed publications.

This question of definition has been belabored here for two reasons. First, the entire community of science has long labored with an inefficient, costly system of scientific communication precisely because it (authors, editors, and publishers) have been unable or unwilling to define primary publication. As a result, much of the literature has been buried in meeting abstracts, obscure conference reports, government documents, or books or journals of minuscule circulation. Other papers, in the same or slightly altered form, are published more than once; occasionally, this is due to the lack of definition as to which conference reports, books, and compilations are (or should be) primary publi­cations and which are not. Redundancy and confusion result. Second, a scien­tific paper is, by definition, a particular kind of document containing specific kinds of information, typically in a prescribed (IMRAD) order. If the graduate student or the budding scientist (and even some of those scientists who have already published many papers) can fully grasp the significance of this defini­tion, the writing task might be a great deal easier. Confusion results from an amorphous task. The easy task is the one in which you know exactly what must be done and in exactly what order it must be done.

2. ORGANIZATION OF A SCIENTIFIC PAPER

A scientific paper is organized to meet the needs of valid publication. It is, or should be, highly stylized, with distinctive and clearly evident component parts. The most common labeling of the component parts, in the basic sciences, is introduction, methods, results, and discussion (hence the acronym IMRAD). Actually, the heading “Materials and Methods” may be more common than the simpler “Methods,” but the latter form was used in the acronym.

Some of us have taught and recommended the IMRAD approach for many years. The tendency toward uniformity has increased since the IMRAD system was prescribed as a standard by the American National Standards Institute, first in 1972 and again in 1979 (American National Standards Institute, 1979a). Some journals use a variation of IMRAD in which methods appear last rather than second. Perhaps we should call this IRDAM. In some journals, details regarding methods commonly appear in figure captions.

The basic IMRAD order is so eminently logical that, increasingly, it is used for many other types of expository writing. Whether one is writing an article about chemistry, archaeology, economics, or crime in the street, the IMRAD format is often the best choice.

This point is generally true for papers reporting laboratory studies and other experiments. There are, of course, exceptions. As examples, reports of field studies in the earth sciences and many clinical case reports in the medical sci­ences do not readily lend themselves to this kind of organization. However, even in these descriptive papers, the same logical progression from problem to solu­tion is often appropriate.

Occasionally, the organization of laboratory papers must differ. If a number of methods were used to achieve directly related results, it might be desirable to combine the materials and methods and the results into an integrated exper­imental section. In some fields and for some types of results, a combined results and discussion section is usual or desirable. In addition, many primary journals publish notes or short communications, in which the IMRAD orga­nization is modified.

Various types of organization are used in descriptive areas of science. To determine how to organize such papers and which general headings to use, refer to the instructions to authors of your target journal and look at analo­gous papers the journal has published. Also, you can obtain general informa­tion from appropriate source books. For example, types of medical papers are described by Huth (1999), Peat and others (2002), Taylor (2011), and contribu­tors to a multiauthor guide (Hall 2013); types of engineering papers and reports are outlined by Michaelson (1990) and by Beer and McMurrey (2014). Indeed, even if a paper will appear in the IMRAD format, books on writing in one’s own discipline can be worth consulting. Examples of such books include those in biomedical science by Zeiger (2000); the health sciences by Lang (2010); in chemistry by Ebel, Bliefert, and Russey (2004); and in psychology by Sternberg and Sternberg (2010).

In short, the preparation of a scientific paper has less to do with literary skill than with organization. A scientific paper is not literature. The preparer of a scientific paper is not an author in the literary sense. As an international col­league noted, this fact can comfort those writing scientific papers other than in their native language.

Some old-fashioned colleagues think that scientific papers should be litera­ture, that the style and flair of an author should be clearly evident, and that varia­tions in style encourage the interest of the reader. Scientists should indeed be interested in reading literature, and perhaps even in writing literature, but the communication of research results is a more prosaic procedure. As Booth (1981) put it, “Grandiloquence has no place in scientific writing.”

Today, the average scientist, to keep up with a field, must examine the data reported in a very large number of papers. Also, English, the international language of science, is a second language for many scientists. Therefore, scien­tists (and of course editors) must demand a system of reporting data that is uniform, concise, and readily understandable.

3. SHAPE OF A SCIENTIFIC PAPER

Imagine that a friend visits your laboratory or office. The friend is unfamiliar with your research and wants to know about it. To orient your friend, first you identify your general research area and say why it is important. Then you state the specific focus of your research, summarize how you gathered your data, and say what you found. Finally you discuss the broader significance of your find­ings. The friend now has a new understanding—and, if you are lucky, he or she might buy you lunch.

Although intended for readers who are more knowledgeable, a scientific paper should take much the same approach: first providing broad orientation, then focusing narrowly on the specific research, and then considering the find­ings in wider context. Some have likened this shape for a scientific paper to an hourglass: broad, then narrow, then broad. Keeping this overall structure in mind can aid when writing individual parts of a paper and integrating them into a coherent whole.

4. OTHER DEFINITIONS

If scientific paper is the term for an original research report, how should this be distinguished from research reports that are not original, are not scientific, or somehow fail to qualify as scientific papers? Some specific terms are commonly used: review paper, conference report, and meeting abstract.

A review paper may review almost anything, most typically the recent work in a defined subject area or the work of a particular individual or group. Thus, the review paper is designed to summarize, analyze, evaluate, or synthe­size information that has already been published (research reports in primary journals). Although much or all of the material in a review paper has previously been published, the problem of dual publication (duplicate publication of orig­inal data) does not normally arise because the review nature of the work is usually obvious—often from the title of the periodical, such as Microbiology and Molecular Biology Reviews or Annual Review of Astronomy and Astrophysics. Do not assume, however, that reviews contain nothing new. From the best review papers come new syntheses, new ideas and theories, and even new paradigms.

A conference report is a paper published in a book or journal as part of the proceedings of a symposium, national or international congress, workshop, roundtable, or the like. Such conferences commonly are not designed for the definitive presentation of original data, and the resultant proceedings (in a book or journal) do not qualify as primary publications. Conference presentations often are review papers, presenting reviews of the recent work of particular scientists or recent work in particular laboratories. Material at some conferences (especially the exciting ones) is in the form of preliminary reports, in which new, original data are presented, often accompanied by interesting speculation. But usually, these preliminary reports do not qualify, nor are they intended to qualify, as scientific papers. Later, often much later, such work may be validly published in a primary journal; by this time, the loose ends have been tied down, essential experimental details have been described (so that a competent worker could repeat the experiments), and previous speculation has matured into conclusions.

Therefore, the vast conference literature that appears normally is not primary. If original data are presented in such contributions, the data can and should be published (or republished) in an archival (primary) journal. Otherwise, the information may essentially be lost. If publication in a primary journal follows publication in a conference report, permission from the original publisher may be needed to reprint figures and other items (see Chapter 19, “Rights and Permissions”), but the more fundamental problem of dual publication normally does not and should not arise.

Meeting abstracts may be brief or relatively extensive. Although they can and generally do contain original information, they are not primary publications, and publication of an abstract should not preclude later publication of the full report.

Traditionally, there was little confusion regarding the typical one-paragraph abstracts published as part of the program or distributed along with the program at a national meeting or international congress. It was usually understood that many of the papers presented at these meetings would later be submitted for publication in primary journals. Sometimes conference organizers request extended abstracts (or synoptics). The extended abstract can supply almost as much information as a full paper; mainly it lacks the experimental detail. How­ever, precisely because it lacks experimental detail, it cannot qualify as a scientific paper.

Those involved with publishing these materials should see the importance of careful definition of the different types of papers. More and more publishers, conference organizers, and individual scientists are agreeing on these basic definitions, and their general acceptance will greatly clarify both primary and secondary communication of scientific information.

Source: Gastel Barbara, Day Robert A. (2016), How to Write and Publish a Scientific Paper, Greenwood; 8th edition.