A simple criterion

Agreement seems lacking, among researchers, on the question of which citations to a scientific article are “legitimate”, i.e., worthy of being included among the total number of hits received by that particular article (typically for the purpose of evaluating one’s h-index).
Should one include a cite to an article, if that cite appears in a manuscript that has not (yet) been refereed, such as a book chapter or a preprint ? What about a conference abstract ? Or, a talk that is available online ?

My simple criterion to discriminate between “legitimate” and “illegitimate” citations is this:
if the source is one that can be cited in a printed journal article, then any citation included therein is a legitimate scholarly citation, and as such it should be treated. It should therefore be included in the total citation counts for all the articles that it cites.

The rationale is simple: if authors, reviewers and editors find a manuscript, or any other source, worth of being cited, regardless of whether it is refereed or not, it is because the information that it contains is deemed cogent and accurate. In that case, there is absolutely no reason not to take seriously the reference list of that very same manuscript or source.
Probably, the way I feel about this subject reflects my views on the value of peer review (see this old post of mine), but I think that this is likely the direction that the community is taking anyway.

In physics, I do not believe that it is presently possible to cite things like abstracts, online talks, blogs, internal university reports, and I am not even sure about theses; on the other hand, ArXiv preprints, book chapters and conference proceedings are regarded as perfectly legitimate sources.

Any objections ?

Tags: , ,

5 Responses to “A simple criterion”

  1. GMP Says:

    I think it’s possible to cite theses (some people say theses count as publications, therefore you are not supposed to present anybody else’s figures in there without explicit permission from their publisher).

    • Isotopic Says:

      Recycling thesis figures is – and has been for a long time – a copyright issue, and theses have always been eligible for copyright protection. With the exception of Fair Use (which applies to any copyright work), or specific uses permitted by their copyright holder (usually the author and the university) figures in theses are just as off-limits as a figure in a textbook or a magazine.

  2. Isotopic Says:

    My understanding is that, in the past, since theses were not widely distributed and somewhat difficult to obtain, citing one was suboptimal because readers could not easily get the original source itself. With the easy availability of digital copies, this is less of an issue, and theses are more suitable for citation.

  3. JF Says:

    Well… I feel this is a byrantine discussion, and in real life it does hardly matter. Is it possible to cite theses? In terms of bibliometry, that makes little difference. If I may take my example, my own thesis collected something like 5 citations, 4 of which where from my own papers to save me lenghtly descriptions (“details can be found in …”). What should we count as “legitimate” citations? Then again, if you are cited 10 times in the “grey” litterature, that’s probably as much as you’ll get.

    In any case, we are talking about a handful of citations, maybe 5 or 10 per item. This is well within the error bar of any bibliometric measurement, as you pointed out in your last post. Or, to put it bluntly: unless you are in very early career, 5-10 citations do not matter on your CV. Either you are reasonably well established, and then nobody will care whether you have 150 or 160 cites, or 2110 or 2120. Or you aren’t, and in that case going from 5 to 10 cites is probably the least of your worries anyway…

    • Massimo Says:

      JF, byzantine or not, the point is not whether citing a thesis is likely to make any significant difference in the computation of the h-index of the person who wrote the thesis– I agree it does not.
      The issue is what to do about the papers that the many theses out there do cite. I doubt very much if the difference to citation metrics that would arise from including their cites is as small as you suggest.
      In some specific subjects (e.g., high-temperature superconductivity) a lot of the theses written in the early 90s cite the same 5-10 papers. If those citations should be counted, those 5-10 papers would see their number of cites probably doubled overnight. Another example is a scientist whose work is mostly methodological — their papers could be cited a lot on theses, and I think it could make a nontrivial difference.

      Furthermore, since we are talking non-refereed publications, it is important that any criterion to include some and exclude others be spelled out precisely. As long as things remain fuzzy, confusion and ambiguity thrive and unfairness to some can ensue.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: