A research ethics article recently appeared that described the results of a survey in which academics were asked whether they had participated in various forms of research misconduct. The responses were organized into specific research misconduct memes, such as 'Fabrication', 'Falsification', 'Authorship Abuse', and others. These all had a short generalized description that seemed typical.

However, there was one category whose description baffled me, meaning that I wasn't sure what they were implying (i.e., what is a typical example in general and would there be typical example in my field (physics).

The category was 'Reference misuse'. While I could think of some practices sleazy in physics that might be called reference misuse, I was otherwise feeling ignorant.

So, what standard example(s) should it conjure up in my mind when 'Reference Misuse' is defined as 'using references to support predetermined arguments rather than illuminate debate'?

(The article that was being summarized the discussed in the blogs is ”Academic Integrity: Exploring Tensions Between Perception and Practice in the Contemporary University” by Joanna Williams and Dave Roberts. It is published by the Society for Research into Higher Education on 30 June 2016). The authors and others report on it in Times Higher education which is where I was reading about it.

  • 2
    I suppose a simple example would be only include the references, which support what ever point you are trying to make. You simply omit the rest, I think this type of behaviour is more common than most people would like to admit.
    – Repmat
    Jul 1, 2016 at 18:50

1 Answer 1


The definition provided seems a little opaque to me, but I can think of a couple of situations to which it might apply.

Reference misdirection: I write a paper in which I include some assertion, perhaps only tangential to my main argument. Maybe it's something that I remember hearing from someone at some point, but I can't quite remember who. The referee or a co-author calls me out on it and says that I need to provide a reference. I type a couple of keywords into Google Scholar and find a plausible-sounding reference. It looks related to my assertion, but it doesn't actually prove it. I cite it anyway knowing that the reviewer will probably not follow up on it.

Overstating the impact of a reference: I write a paper that depends crucially on some Theory X. There is a paper that provided some evidence in support of Theory X, but I cite the paper as claiming that it found "strong evidence in favor of Theory X". Or perhaps the paper did indeed find strong evidence in favor of Theory X, but I state that it "proved" the validity of Theory X. Moreover, perhaps the paper provided a number of qualifications to their claims (which maybe invalidate its application entirely to my case!), but I ignore these qualifications entirely.

Presenting a one-sided history: I write a paper in favor of Theory X (or maybe my paper depends on the validity of Theory X). In my introduction, I provide an extensive survey of the field and cite dozens of papers that have found evidence in favor of Theory X. However, there are just as many papers which have found strong evidence against Theory X, but I do not cite any of these. Although there is clearly a lively scholarly debate surrounding Theory X, I present Theory X as though it is a settled fact. While this won't fool anyone who does serious research on Theory X, it will be a source of confusion for those who work in other fields and are not aware of the controversy.

This last one is particularly common and isn't as egregious as the others since the reference is not fraudulent, just superfluous.

Scratching someone's back: I've written a paper in a related field to a colleague. I know that they would appreciate a citation to one or more of their papers. Or perhaps I would like to cite them to help bring my work to their attention. Their work is not directly related to mine, but I figure out a way to add a few sentences to the introduction that vaguely connect my paper to their work so that I can work a citation to them into the paper.

Since researchers are all human (and hence biased and involved in academic politics), it's impossible to be entirely immune from all of these. If you do a lot of research on Theory X and believe it to be true in your heart of hearts, you will be more aware of the limitations of studies purporting to disprove Theory X and will be more easily able to justify omitting references to them in your work. But the goal of research is to discover the truth and present it as clearly as possible, and part of that process is being as aware of one's own biases as one can.

  • I'm still having a bit of a hard time with this. The "presenting one-sided argument" doesn't seem to translate as well into the physical sciences except. My oversimplified view for most of the papers I read or review in materials physics, the authors data/derviation/theory discussion supports the authors argumemts, and the other stuff tends to be either 'fluff' or 'boilerplate' in the 4-6 page papers that are typical. I was hoping get better insight on this one from academics who are from fields where citing arguments from literature is the major 'data' used.
    – Carol
    Jul 4, 2016 at 15:44
  • But - upon reflection, we are missing some examples under Reference misdirection. This category is the most likely to occur (or at least it is the only that I remember encountering). So if @Joe_Antognini would also include the following 1) Not citing ones own prior work to try to hide or obscure that similar results/figures/paper have been published by the authors. The example in the answer is more clearly stated as 'not reading citations that you have put in your paper' which is very sloppy, but not necessarily fraudulent. (But still should stay in this example, of course)
    – Carol
    Jul 4, 2016 at 15:46
  • And lets ignore pseudoscience papers, which tend to be obvious examples of presenting one-sided arguments and mis-citing arguments from literature as their major 'data' arguments.
    – Carol
    Jul 4, 2016 at 16:48

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .