Science is, among other things, a communal collection of knowledge. Since the growth of scientific knowledge builds on past knowledge, it requires mechanisms for preserving and disseminating knowledge within the scientific community.
Scientific publications are the primary tool for this essential activity of science, and a primary reason for the existence of professional societies like SPIE.
New publications are placed in the context of the communal collection of knowledge through the use of citations, defined as a reference to a source of information or data. Citations in a scientific paper serve many goals, but there are five important goals.
- Provide sufficient context of the work to allow for critical analysis by others, and thus to enable the readers to gauge for themselves whether the author’s conclusions are justified
- Give the reader sources of background and related material so that the current work can be understood by the target audience (thus creating a web of science)
- Establish credibility with the reader (e.g., the author knows the field, has done his/her homework, etc.) and/or inform the reader that the paper belongs within a specific school of thought
- Provide examples of alternate ideas, data, or conclusions to compare and contrast with this work
- Acknowledge and give credit to sources relied upon for this work (i.e., acknowledge the use of another’s ideas or data), thus upholding intellectual honesty.
While most authors do a good job of providing citations in their scientific papers, some papers are flawed because of missing citations, i.e., not enough citations to fulfill the five goals above. Other problems include references that are not needed but are included anyway (spurious citations) and ones that are added or omitted for reasons other than meeting the five goals of citations (biased citations).
In general, authors can mitigate these citation problems by asking two questions:
- Have I provided the references that will make this paper as useful as possible?
- If the reader looks up a given reference, will his/her time be well spent?
BE CAREFUL WHAT YOU MEASURE
Additionally, citations sometimes have significance for reasons other than the five listed above. They can be counted, and in a data-driven world, these counts have assumed outsized importance as a proxy for the influence of a given paper. Citation counts also serve as (flawed) measures of journal importance (the impact factor) and researcher clout (the h-index).
Today, such citation counts and their metrification are used in hiring and promotion decisions, especially in academia, often as a substitute for thoughtful and informed judgment. They also influence research funding strongly in some parts of the world.
Be careful what you measure, since a truism of the business world is ‘what gets measured gets managed.’ And measures that come with rewards often get gamed. When a person’s career or reputation depends on citation counts, the temptation to inflate those counts is never far away.
Some authors are more likely to cite their colleagues’ work than their competitors’; some journals expect their submitting authors to preferentially cite work published in that journal. However, the easiest way to promote your own work (and thus yourself) is with the self-citation: a citation to one’s own prior work.
SELF-CITATIONS IN CONTEXT
Self-cites are not inherently problematic (notice my self-cite in this article). Most scientific publications describe a part of a longer-term research effort, and self-citations can put the new publication in the context of that larger effort. Self-cites become a problem only when they are either spurious or biased.
Since deciding that a specific citation is either spurious or biased requires a judgment based on the cited work, the paper in which the citation occurs, and the field within which the work resides, it is not an easy evaluation to make. Some cases are obvious, as when a majority of a research paper’s citations are to the author’s own work in a popular field of research.
Other cases are less obvious, as when the authors are nearly the only ones working on a very specialized topic. Still, I think most authors know when they are pushing into spurious or biased territory with their self-citations. So the best defense against abuse is self-regulation.
Or is it? A study commissioned by the Chronicle of Higher Education looked at the role of gender in self-citation rates. An examination of 1.7 million JSTOR papers spanning across disciplines and over 60 years found that nearly 10% of citations were self-citations.
WOMEN CITE THEMSELVES LESS
Further analysis showed that men were 56% more likely to cite their own work than women were, with the gender disparity growing over time. Apparently self-regulation of self-citations is more effective in women than men.
What is the cause of this gender disparity? Women in academia seem less inclined to self-promotion than men, probably to their detriment. Does society pressure women to be more “feminine” and modest about their accomplishments? Are men encouraged to be more aggressive in pursuit of career success? Do women work on smaller teams with fewer publications and fewer opportunities for self-citations?
I am certainly not qualified to address such heady questions, but regardless of cause, the issue of gender disparity in self-citations has career consequences.
In the age of Big Data, success breeds success, and popularity snowballs. The most-linked web pages, the most-watched videos, and the most-downloaded journal papers are recommended or promoted to website visitors and social media consumers, generating a handful of winners-take-all and a long tail of neglected also-rans.
The bandwagon effect seems true in the world of academic citations as well. Could it be that even modest differences in self-citation rates might snowball into noticeable differences in total citations? In other words, does self-promotion through self-citation work?
One 2007 study reported in Scientometrics showed that it does, with each self-citation multiplying into three other citations to that author over a five-year period.
DRAWBACKS FOR YOUR READER
Further, the penalties for excessive self-citation seem to be small or none. While this 2007 study looked at papers published from 1981 to 2000, I imagine that the higher levels of online searching and reading today have only increased this multiplying effect. Differences in self-citation rates are likely only one of many factors contributing to gender disparities in academic careers. But it may be one of the easier ones to address.
Proper citing requires careful consideration of the appropriate goals of citations, aided by a simple ethos: make the paper reader-centric, not author-centric. Self-promotion is an author-centered way of looking at the activity of publishing and is neither good nor bad when considering the needs of the reader.
While self-cites should not be added to a paper solely for self-promotion, neither should self-cites be avoided for fear that they might appear self-promoting and unseemly. By focusing on the reader and the five proper goals of citations, you can easily avoid most problems concerning citations.
- Chris A. Mack, “Editorial: How to Write a Good Scientific Paper: Citations,” Journal of Micro/Nanolithography, MEMS, and MOEMS 11(3), 030101 (2012).
- James Hartley, “To cite or not to cite: author self-citations and the impact factor,” Scientometrics 92(2), 313–317 (2012).
- Robin Wilson, “Lowered Cites,” The Chronicle of Higher Education 60(27), (March 17, 2015) http://chronicle.com/article/New-Gender-Gap-in-Scholarship/145311/. Accessed 7/2/2015.
- James H. Fowler and Dag W. Aksnes, “Does self-citation pay?, Scientometrics 72(3), 427–437 (2007).
– SPIE Fellow Chris A. Mack is editor-in-chief of the Journal of Micro/Nanolithography, MEMS, and MOEMS and frequently writes about scientific writing, lithography, and Moore’s Law. The recipient of the 2009 SPIE Frits Zernike Award for Microlithography, Mack founded FINLE Technologies, the developer of the lithography simulation software PROLITH, in 1990 and served as its president and CTO until the company was acquired by KLA-Tencor in 2000. He has a PhD in chemical engineering from University of Texas at Austin (USA) where he is an adjunct faculty member.