Introduction

One of the most important, yet difficult things for students to learn is how to find, read, and interpret various forms of scientific literature. Professors send you off on various assignments, for a class or research project, and urge or require that you use "primary" literature. Students struggle to find appropriate sources to read – both for their own understanding of the questions/issues at hand, as well as to "cite" as literature used to meet some assignment. I would like to take a few moments, here, to place some perspectives on the various forms of literature that are available and how one might use them. I hope that my comments will help you come to better understand what is meant by "primary literature" versus other sources of information.

Peer-review and what it means in Science

One of the most important, and misunderstood, concepts in science is that of peer-review. When we say that a work (magazine article, book, grant proposal) is peer-reviewed, we mean that it has been submitted to a rigorous, critical review by professional peers prior to its publication or granting of funds. Typically, there is a formal mechanism whereby a panel of experts in the field receive a work and solicit critical reviews (anonymous to the original author). The purpose of such pre-publication reviews is two-fold, 1) to ensure that the work does not contain serious flaws in methodology or logic that would be rejected easily by the scientific community upon publication, and 2) to ensure that the study or ideas are clearly described so that the scientific community can critically evaluate it upon its publication. So, when we speak of "peer-reviewed publications" we mean those that have met a standard of pre-publication review that position them for evaluation by the professional scientific community. Later, after publication, peer-review takes on another meaning. As I discuss below under another heading (The "test of time"), a researcher’s ideas are challenged and evaluated by their peers.

Primary sources – peer-reviewed journals

Primary journals (e.g., Science, Nature, Journal of Experimental Biology, etc…) publish articles based on original scientific investigation. Most journals focus on original articles (or reports), but may also occasionally publish special invited review articles. Invited works are usually identified as "reviews" or sometimes they are identified as "perspectives" or "forums". They have not usually undergone the same level of peer-review as original articles, as they are typically aimed at summarizing or reviewing a major field, rather than presenting new results or interpretations. They typically are authored by researchers with reputations as experts in the field and have already established themselves as people who can and have produced many fine works that have met a high level of peer-review. Perspectives and forums are usually written with an intent to put forward a definite position to challenge an idea or interpretation and often involve a pair of authors who take different viewpoints. Again, like reviews, these works are usually not rigorously peer-reviewed and thus, arguments made are more provocative than based on data.

Most primary journals have "book reviews", "letters to the editor" and/or "corrections/addenda". Letters to the editor give readers an opportunity to comment (usually very briefly) on papers that have been published in the journal, typically focused and directed at the methods used or noting some important literature that was not cited in the original article. The original authors are usually given an opportunity to respond. Such letters give you a sense of the kind of thing that goes on during the peer-review process itself. Corrections/addenda are noting mistakes made in the printing (maybe a figure was mislabeled) or analysis that was done improperly.

Recently, some major journals (see esp. Science and Nature) have developed a section of news summaries for selected original articles that appear later in the back of the journal. These summaries are written by science-writers or by other scientists playing that role. They are intended to provide interested readers (including even scientists!) with more accessible descriptions of a primary journal article. They are not peer-reviewed, and in that sense, shouldn’t be considered original work. However, these articles often describe the research in ways that make it understandable to the non-specialist. That’s why professors pass these out to their students (I’ll give you several this semester). You should certainly use these to understand primary literature – just realize they are not primary literature.
 
 

Secondary sources -- Review and Educational Journals

In addition to what I’ve been describing as primary journals, there are many that hold a special place as review or educational journals. These publish exclusively invited reviews and or summaries of primary literature. Examples of review journals include, Quarterly Review of Biology, Cambridge Biological Reviews, Biological Review, Trends in Ecology and Evolution. There are also specialized series such as the Annual Reviews of Ecology and Systematics (and some 20 others), International journal of Cell Biology, and so on. Journals I refer to as educational journals include, American Scientist, BioScience, and Scientific American. These are specifically aimed at providing educators with summaries of current ideas in various fields.

Review and educational journals contribute many functions, including supporting general understanding of ideas and methods, reviewing the body of literature in a field, and serving as a "jumping-off" point for new researchers to pursue questions in a given field. The new researchers I’m referring to are not only students (undergraduate or graduate), but seasoned faculty as well. Many of us professorial types use reviews and educational summaries to "catch-up" on fields that we are not closely following and/or to enhance our general knowledge for teaching a subject. These articles (and journals) are also very valuable for undergraduate students. Quite often, these are the articles that your professors distribute to you in undergraduate courses and they also can be researched to help you write a paper. So, one should not view them as somehow "inferior" because they are secondary, rather than primary, sources. However, if a given assignment requires you to find primary journal articles, then you need to find them and these review articles are a good place to start ! If you can find a good review article on a subject, you can read it to gain a general understanding of the major questions and approaches to a problem, then search-out the primary articles referenced in the review article.
 
 

What about textbooks ?

The textbooks you’ve been using for many years now are not primary sources, although you and your instructor probably treat them as such. They really are secondary to tertiary sources, telling the "stories" of how biology works in a declarative manner. Most have few references and even these are usually not primary sources, themselves (check out the references at the end of a chapter and note the number of books and review articles, as opposed to specific journal articles). Texts are written by people who have been or are researchers, but who have a deep commitment to teaching. It takes hard work to write a textbook. It requires the same kind of researching that goes into writing a major review article. It also requires bringing in some presentation techniques (good illustrations, etc…) that are usually outside of what is allowed in primary journal or review articles. You can always refer to textbooks (the one you’re using in this course or others), as a very broad starting point, but do not stop there and do not use these are primary (or even secondary) sources.

What about other books?

In biology (and science, generally), the highest form of scholarship is publishing papers in the world’s best primary source journals. The "best" are defined as those with the highest level of peer-review prior to publication and those that will have very wide dissemination (will be read widely) and therefore subjected to the highest level of post-publication criticism. So what about books? By books, I mean scholarly books, usually published by academic or independent presses with an intended audience of academics (e.g., Academic Press, Plenum Publishing, Cambridge Univ. Press, Oxford Univ. Press, Univ. of Chicago Press, etc…). Most of these presses have editorial review boards of researchers that do, in fact, review the manuscripts before publication. These reviewers are often the same people who would be peer-reviewing primary journal articles on the same subject. One should perhaps best consider academic books as equivalent to primary articles presented in a very long (extended) format. The long format gives the author the opportunity to develop ideas at length and to present more data and interpretations than is usually tolerated in journals. For most undergraduate students and course assignments, books don’t make a great source for research. This is because the work essentially represents a single source -- a single person’s experiments/observations and interpretations – even though the work is very long and developed in detail. You should use these as general references or to gain access to the literature. Of course, over time you read all the important books and have them as part of your intellectual corpus.

Another form of book that is very common is the edited volume. That is, a book with many separate "chapters" written as separate articles by different authors – one or more "editors" serve as organizers of the entire volume. Edited books are often the result of a scientific conference or symposium. These works are sometimes extremely rigorously peer-reviewed and containing outstanding scientific work, and sometimes not. They are rigorous, high quality works if the editors have exercised a peer-reviewed structure, but aren’t if the work is basically one in which the authors are given a free hand to write anything they desire without review. It’s hard to know which is the case, except that (generally) professionals in the field know who does a good job as an editor and who doesn’t. These works can and should be found, read, and used in one’s research, but should always be evaluated with care.
 
 

Scientific/Nature Magazines for the "educated public"

Finally, there are public magazines (including newspapers). These are not peer-reviewed in the strict sense, but do have editorial staff who seeks to evaluate them for content and editorial style. Biological magazines like Natural History, Smithsonian, Audubon, National Geographic and others, fit in this category. Truly public sources like Time, Newsweek, U.S. News & World Report, and daily newspapers are published works, as well. These latter sources are usually at least secondary, if not tertiary or quaternary forms of information (that is, they are tell the "story" at least 2 levels beyond the original source). Generally speaking, these are not the places for you to gain a general understanding of some scientific concept (they are not rigorously scientific or academic) nor are they a source that can help you write a research paper. However, some biological magazines do have carefully researched pieces (although usually without references) by good researchers (e.g., Jared Diamond’s or Stephen J. Gould’s essays in Natural History). Again, professors sometimes distribute these to students to provoke some thinking or understanding. They are not primary sources, so they shouldn’t be treated as such.
 
 

The "test of time"

I’ve been commenting on peer-review as a process of critical assessment that evaluates works prior to their publication. However, there is an even more fundamental aspect that usually gets the lion’s share of explanation to students. Which is, the review process that occurs as people read a work and then criticize it. Ultimately, science progresses as people repeat experiments, observations, or calculations to test published ideas. One might even say that the peer-review process that goes on before an article or book is published is irrelevant to the review process after publication. The reason why peer-review prior to publication has become so important, is the highly specialized and voluminous nature of research today. Hundreds of thousands of scientists are pursuing studies and trying to disseminate their results annually. Peer-review during the publication process provides some initial evaluation of the scientific merit of a study or idea, and of its being clearly described for the professional community. The bottom-line, though, is to recognize two statements that often ring true -- 1) just because something is published doesn’t mean it’s right (even in the present), and 2) some very good work is published in the smallest (most obscure) publication outlets. It means you have to constantly search for literature (especially outside of electronic databases that miss the small journals), and don’t believe everything you read.

Delivery by the Internet

It has come time to address some interesting new "problems" that have evolved as the world wide web has connected our world. The www is a fantastic tool to connect people and their works. It is dynamic and allows remarkably open access to people and their works. Anyone can generate their own work and "publish" it to the web, for others to use. Publishers are scrambling to figure-out how to put their publications on the web, yet insure that their copyright is protected and that they can make a profit in the process. Nearly every major primary journal (and review/educational journals), as well as more public publications have web sites that minimally allow access to the table of contents. Most journals are now allowing their subscribers (or members) to have full internet access to the articles. We should view this as simply providing a more dynamic, convenient method of delivering the journal to a reader. Listed below are several major sites you can visit, to see how this works.

It has become obvious to me and many of my professorial colleagues, that several "problems" have arisen with this new delivery system. [Please note that I am writing this commentary "on the fly" without references and without the intention of doing academic research on this topic – there are many, many high quality references on the topics I am addressing here].

1. The problem of documenting/citing sources. The Internet is extremely dynamic, you can view, save, and even copy materials from a site without including an attached documentation of its source. We approach the www so informally, that student and faculty researchers who wouldn’t dream of not giving complete citation or attribution of a work to an author/institution, ignore documentation of source without blinking an eye. This informal approach to the web and its materials would be fine if our use of Internet-delivered work was limited to self-understanding or entertainment. But it’s not! We are using the web as a major delivery mechanism of knowledge. It’s not going too far to say that this medium is as revolutionary as the printing press was over 500 year ago. Therefore, despite the web’s origin as an informal form of communication and our causal approach to it, we must be vigilant on treating materials delivered by the web as sources needing documentation.

2. The problem of identifying works as peer-reviewed. – As more and more journals place their materials on the www, (see esp. Science and Nature), it is becoming harder and harder to distinguish those works from other sources. Moreover, even "prestigious" journals are sending mixed messages. The various kinds of articles (original reports, reviews, perspectives, forums, letters to editor, and public summaries) are not always clearly differentiated in the online versions of the journals. For example, in Science, the peer-reviewed original articles (they are called reports) are available as web-based abstracts and also pdf-formated copies. The pdf copies are exact copies from the journal pages. Most of the letters to editor, public summaries are in web-format only.

How are we to understand which works are peer-reviewed and which are not? We need to come to understand the original works (i.e., the printed version of a journal) and the relationship the various articles published there have with the online versions of the same publication. We need to work hard to make sure we know the ultimate source of all web-delivered materials and maintain their documentation.

3. Doing "searches" captures a lot of stuff.—Going on the Internet and simply doing a search captures a ton of material – some are primary journals or journal articles, others are secondary sources, others are tertiary, public, or other sources, and finally others are from unknown sites. The unknown sites are usually people, including professors with class materials on the web, researchers or students who are interested in "creating" work for themselves or others. Some of these materials are useful, others are not. Little or none of it is adequately documented (to original or even secondary sources). To put this in scientific terms, most web material gives us no clue to proximate, let alone ultimate causes (sources). Since I’m placing this document on the web, someone out there in the world will likely find it and, thinking it’s "pretty good" will use it in some way (linked to their course website or copied to their students).

What to do ?

To find, read, and use information in our information age has been made both easier and more difficult with delivery by the www. We can access information much easier, but have a much harder time identifying its source, knowing the level or its review (peer-review by colleagues in the field versus a copy editor versus none). To use information today means we have to be more vigilant than ever in understanding and documenting source information. Here’s my own checklist of things to remember:

1. Clearly identify and document./attribute the authors of a work

2. Clearly identify and document the organization/publisher of a work

3. Clearly understand and document the level of peer-review applied to a work.

    <= Return to Dr. Hood's Homepage