The Anatomy of a Scientific Paper
Adapted from an assignment by Thomas Holtz. (Thanks, Tom!)
Now for something completely different:
While the hypothetico-deductive method is the heart and soul of science, the transmission of research resuts is how Science grows. The principal means of transmitting scientific work is the scientific (or technical) research paper. These are normally published in periodical journals (which may come out weekly, monthly, or on a longer schedule). Each paper is a discussion about a particular new discovery and analysis, giving the background information, what was studied, and the results. Let's see how they are organized.
Access for students:First, a word about accessing the technical literature. While some journals (such as the Public Library of Science series, http://www.plos.org/ ) are freely available to everyone (i.e. "Open Access"), most journals (and even their electronic versions) require paid subscription. The good news is that YOU don't have to pay for them (directly), since the University has subscriptions to thousands of these. But getting to them may be a bit of a pain until you learn the drill.
- From on campus (including the campus wireless system), it is actually easy: the journal's website recognizes you accessing it from a subscriber. But off campus requires a couple of different approaches. The quickest way is to change the URL (web address) of the journal article by adding the following text into the address manually (cutting it and pasting it after ".org" or ".com" or whatever):
So if you want to read the article http://www.sciencemag.org/content/345/6204/1613.abstract while off-campus, change it to http://www.sciencemag.org.proxy-um.researchport.umd.edu/content/345/6204/1613.abstract
From here you get a sign-in screen, which double checks that you are indeed a UMd student.
- Alternatively, you can use the following URLs for search engines for which the University has a paid subscription:
- http://apps.webofknowledge.com.proxy-um.researchport.umd.edu/ or
If you encounter a situation where the site is asking you to pay for an article, stop! We do not want you to pay for any article that you should get to for free.
The Papers Themselves:There are many publication formats: fiction, essays, reports, blogs, press releases, and so on. One of the most important for Science is the technical research report (aka a "scientific paper"). This is the sort of paper, typically published in a regular journal but occasionally in an edited book or other volume, where a scientist or team of scientist put forth their discoveries, ideas, hypotheses, tests of their hypotheses, and conclusions.
Consider a recent one:
Long, J.A., E. Mark-Kurik, Z. Johanson, M.S.Y. Lee, G.C. Young, Zhu M., P.E. Ahlberg, M. Newman, R. Jones, J. den Blaauwen, B. Choo & K. Trinajstic. 2015. Copulation in antiarch placoderms and the origin of gnathostome internal fertilization. Nature 517: 196-199. https://www.nature.com/articles/nature13825?message-global=remove
A few notes:
- This is a technical paper: its intended audience is not undergraduate non-majors, but other professional scientists. Do NOT expect to understand it fully!
- I listed this reference in the CSE (Council of Scientific Editors) bibliographic format that I prefer. You can find more about this at http://www.geol.umd.edu/~tholtz/G204/204bib.html. Please familiarize yourself with it.
Whether they are labeled or not, most technical research papers contain the following sections:
- Title: sometimes a straightforward description of the work, sometimes a literary reference, sometimes a pun.
- Authors and Affiliations: All the authors who contributed to this paper. Sometimes just a single author, sometimes a few, sometimes literally hundreds. Additionally, the affiliations (what institutions they work for) and addresses of the authors are given. Normally one particular author is indicated as the "Corresponding Author", and their email address is provided if you want to contact them.
- Abstract: often a 1-2 paragraph statement at the very beginning of the paper identifying the major observations, discoveries, and conclusions to be discussed. You read this to decide whether you are interested in the rest of the paper.
- Introduction or background: what the issue at hand is, previous work, perhaps a synopsis of competing models about the topic.
- Materials and methods: what things were studied, what measurements were taken (and with what instruments), what analytical techniques were used, what equipment was utilized. Typically presented between the Introduction and the Results; however, with the recent (last decade or so) rise of the "Online Supplementary Material" some journals have shoved Methods & Materials to the end of the paper or entirely off to the online supplementary material. If the latter, there should a statement about this in the main paper. Wherever it may be, it provides the essential information to allow someone else to replicate the work.
In paleontology and organismal biology, the "materials" may consist of a new organism whose description takes up most of the section.
- Results: when the author(s) used their methods on the materials, what did they find? What were the results of the analysis? This section will almost always include at least one chart/plot/table or other graphic representation of analytical results.
In papers devoted to the description of a new organism, the results section is reduced or absent.
- Discussion: what are the implications of the results? Do they support a particular hypothesis on the subject, or reject it? What new aspects about the phenomenon in question were revealed, if any?
- Future work: not always present in every paper, but there is often some statement about the potential of this analysis to lead to some new as-yet-not-done analyses, pointing the way for future research.
- Conclusions: a summary of the major points of the analysis.
- Acknowledgements: a list of people and institutions who aided with the completion of the projects discussed. If the study was funded by a grant, the granting agency and number and name of the grant is typically listed here.
- Bibliography: a list of references cited in the work. You will see that different journals have different bibliographic styles.
Take a look at the Long et al. (2015) paper mentioned above and see if you can identify these. (The abstract isn't directly labeled, but it is in bold and is at the top of the paper.)
Varieties of Scientific Publications:There are a number of types of articles in science beside technical research papers. Some are published in the very same journals as the scientific research papers, and some in different venues. Here are some of the major types you may encounter when researching information. You must be able to distinguish them:
- Review papers: these often have the same general properties as a scientific technical research paper with one major exception: they are not announcements of new discoveries nor new analyses of data. Instead, reviews are summaries of the current state of research in a particular field. They typically do not give new analytical results or tests of hypotheses. However, they have an important role as places where different papers (often by many different researchers, perhaps but not necessarily the present one) are synthesized into a whole. They are also extremely useful in finding the original research (as references) in a particular field. Review papers are sometimes included within the typical scientific journal (and if so are normally indicated as such). Additionally, there are entire journals (e.g., the huge line of Annual Reviews journals; the Quarterly Review of Biology; etc.) which are entirely review papers.
- Conference abstracts: these are "proto-papers" in a sense: one-to-two paragraph overviews of the work presented at a technical conference (either as a stand-up platform presentation or as a poster). Often these are the seeds of a future publication. But because just the abstract is presented, and because neither the data nor the analyses yet been subjected to peer-review, they are not regarded as "finished works" in the scientific community. They are occasionally cited in technical research papers because they serve as documents as to when a particular idea or discovery was first presented to the scientific world. However, when the full research article is published, it is that rather than the conference abstract which is regarded as the "real" analysis.
- Commentary: these are labeled as such (or with a special name, such as the "News and Views" section of Nature or the "Perspectives" section of Science). These are normally reviews of a particular technical paper (or perhaps multiple papers) in that issue of the journal, in which a different expert in the same subject is asked to present the important broader implications of that technical paper. (In contrast, true review papers review of the state of the field as a whole rather than a subset of particular papers.) In addition to the having a special designation, you can recognize a commentary by:
- having slightly less "jargon-y" text, as it is there to translate the highly technical language of the original article into something a non-specialist might appreciate, and;
- by indicating (typically in the first two paragraphs) the specific article(s) later in that issue which are being discussed.
- News reports: many journals will give short reports announcing the publication, results, and implications of significant new technical publications. These might be papers later in the issue of that journal, or they might be in entirely different journals. News reports can be distinguished from technical research reports (and indeed all the over types of articles already discussed) in that they:
- often have no specific author listed;
- are very short (maybe just a single paragraph);
- normally have no references, except for the journal article that is being reported;
- do not show any actual analysis.
- Editorial: some--although by no means all--scientific journals may have an occasional editorial or opinion piece. These are indicated as "Editorial" or "Opinion", and consist of an editor's (or invited author's) thoughts, comments, and opinions on some pressing topic. These almost never have new analytical results, and will typically only have a few references (compared to an actual research article). These are useful to find out what particular editors think about various policies and subjects, but are not scientific results as such.
- Press releases: these might be from the home institution of a researcher, or a granting agency, or a corporation, or other protagonists. In any event, it is typically from an institution announcing some breakthrough by one of its employees or beneficiaries. (This is for a number of reasons, not least of which is to justify to the shareholders in that institution that their money is being put to good and productive use!) Note that these press releases are not subject to independent peer review, and consequently are not considered valid references by the scientific community. Their primary value is to communicate with the media and the general public. Similarly public or private institutions and corporations may put out press releases on particular scientific topics. They might be useful in terms of presenting data not otherwise available (e.g., the statistics and measurements of a particular piece of scientific equipment, for instance), and so might be cited in that context. However, analytical results and observations in these are again not subject to independent peer review, and are thus potentially suspect.