Skip to main content
It looks like you're using Internet Explorer 11 or older. This website works best with modern browsers such as the latest versions of Chrome, Firefox, Safari, and Edge. If you continue with this browser, you may see unexpected results.

BSCI 1510L Literature and Stats Guide: 2.3 Traversing the web of citations

Introduction to Biological Sciences lab, first semester

2.3 Traversing the web of citations

After using Web of Science for a while, you will get a feel for how the scientific literature system allows a researcher to explore what is known about a topic by examining earlier papers that are cited in a particular paper or later papers which cite the particular paper.  Clicking on the Web of Science "Citation Map" link illustrates this graphically.  The availability of electronic databases which contain citation links and the text of abstracts has made this process much easier.  Using tools like Web of Science allows you to examine the abstracts of other papers which are linked through citations and assess whether that paper is likely to be relevant without requiring you to actually look up the paper and read it.  This makes it possible to start with a paper that is only marginally relevant and to find your way to other papers whose topic is closer to the subject you want to investigate. 

An important class of journal articles that we have not previously discussed is a review paper.  A review paper is classified as a secondary source of information, because the purpose is not to share the results of original research, but to present an organized, evaluative summary of the important publications on a particular topic. A simple hack for distinguishing between an original experimental paper and a review paper is to look for a “methods” section. In a primary source, the authors themselves participated in the described research; therefore, if a scientific paper has a methods section, it is likely reporting original research.  If you are lucky enough to find a recent review paper on the topic that you are researching, you may be able to get "up to speed" on the literature much more quickly.  There is also a group of series called Annual Reviews, which each year publishes a volume of review papers summarizing important recent developments in the field covered by that particular series (e.g. Annual Review of Ecology, Evolution, and Systematics, Annual Review of Biochemistry, etc.). 

If you are reading a paper online, you can sometimes jump directly to a cited paper by clicking on a link.  However, an obvious limitation of using the reference section of a paper is that it can only cite papers that were published at an earlier time.  This is particularly problematic if the paper itself is somewhat outdated since you may be missing very significant papers that were published recently.  Fortunately, Web of Science can show you papers in their database which have cited the paper you searched for.  In the results page for any search hit, click on the numbered link which follows "Times Cited:".  This will take you to a "Citing Articles" summary page which lists the articles which cited the paper for which you searched.  Clearly these will all be papers published after the article for which you searched.  The authors of those citing articles thought that the paper you searched for was relevant enough to their work to cite it, so there is a good chance that at least some of these citing papers will be helpful to you.  By examining the abstracts of those papers, you can determine whether it would be fruitful to actually look them up and read them. 

It is also possible to find papers that cite a work which does not itself have a record in Web of Science.  For example, Ph.D. theses are not themselves indexed in Web of Science.  Click on the Cited Reference Search link, then conduct a search for Cited Author= baskauf cj .  Check the box for "THESIS VANDERBILT U" and click "Finish Search".  The results page will show the articles which cited the thesis. 

PubMed also provides a citing articles listing.  On the results page of a search hit, the right side of the page lists "Cited by X PubMed Central articles".  However, this list is likely to be less comprehensive than Web of Science, since it only includes papers that are freely available as full text. 

Google Scholar (Section 2.4.3) shows citing articles on their search results page.  Under each hit there is a "Cited by X" link which will take you to another page of search results containing papers which cited the hit.

Judging the quality of sources: what makes a scientific journal "reputable"?

Do I have to use recent sources? How recent is “recent”? There is no one straightforward rule for how recent a source should be. Generally, use the most recent and updated sources as possible to support your paper, preferably those published within the last ten years. However, here are some general exceptions to this rule:

  • Introducing a paper for its historical significance. For example, if your paper describes the discovery of DNA’s structure, and you want to cite Watson and Crick’s original 1953 paper, this would be an appropriate historical citation.
  • To credit whoever designed an experimental procedure in your methods section. If you use a procedure that was created by another scientist, cite the paper that introduces this method. It would be fine to cite a method that is older, if it is still commonly used today.
  • If you are writing about a very niche and rarely studied topic. For a topic that is narrow in scope or poorly funded, you may find it impossible to find a review article from the last five years, and will need to cite sources slightly older. The more commonly studied your topic is, the more recent your sources need to be. As an example, if you are writing a paper on cancer biology, do not cite sources from 1989, because there are thousands of papers that are published on this topic every year.

Journal Impact Factor (IF). Journal Impact factor is a numerical score assigned to journals based on the yearly average of how many times their articles are cited by other publications. An impact factor of 5 and above indicates that the journal is in the top 6.5% of published journals. This is a useful, but imperfect measure of the reputation of a journal – it is not unusual for poorly designed or poorly written articles to appear in high-impact journals, and strong research may be published in journals that are new or have little name recognition. IF is most useful when evaluating the overall sources of your research. For example, if you are writing a research paper on a controversial or unfamiliar topic, and you find that all of the journal articles you are citing have very low impact scores, this may indicate that you are “cherry-picking” studies that are not supported or reproducible.