16

In my lab, a few of us submitted our papers independently to a conference last week. We all had referenced each other's work in our papers. I have a concern if this is a bad practice in academia. It also looks like a bad habit that has festered in our lab: the last two PhDs who passed out from our lab have a total of 15 citations for their papers, of which 13 are from within our community itself.

Is this practice of repeatedly citing from one's own micro-community acceptable? It seems unavoidable, as we are all working together, familiarizing with each other's work and trying to build on them.

|improve this question
  • 2
    I changed the title, as a number of users expressed concern that "suicite" isn't the appropriate term to use here. Please verify that the new title is appropriate. – eykanal♦ Jun 2 '13 at 2:38
  • 1
    How recently were those two PhDs graduated? I suspect "organic" external citations appear much more slowly than internal citations from people who are both working in the same area, and already know of the author's work. You all get to skip the "discovery" step. – Fomite Jun 2 '13 at 7:09
  • @eykanal: The title is alright, but I would like to state "suicite" is etymologically driven and has no malicious intent. – Sara Jun 2 '13 at 9:49
  • @EpiGrad: One of these PhDs graduated last year and another this year, each after six years in PhD. I am in my third year myself. – Sara Jun 2 '13 at 9:51
  • 2
    I find the question confusing. Can you edit it to be more explicit? Are you saying that most of the references in your recently-submitted papers are to each other? When you cited each other's papers, were those papers as relevant as everything else you cited? Would you have cited them, if they weren't from your group? When you mention the 13 of 15 statistic, does that mean that they wrote a paper/dissertation where 13 of the 15 citations are to papers in their own team; or do you mean that only 15 other papers have cited their work, and 13 of those citations come from within their own team? – D.W. Jun 4 '13 at 19:21
14

It is only bad practice if the work is not particularly relevant, or if you cite your own group's papers instead of other relevant work. The point of citations is to lead the reader to other related work necessary for some aspect of the paper (result, context, etc.).

That said, it is worth examining why those papers aren't getting cited. Maybe nobody else is working in a sufficiently related area but it's critical background for the rest of what the lab is doing--then I wouldn't worry about it. But if other people are working in the area (e.g. to the point that you might worry that if you take too long that someone else will publish the same results you're trying to get), it can be a sign that the results either aren't contributing to the field much, that the style of presentation isn't good enough for others to really understand what you're saying, or that the lab members aren't going to enough conferences to present.

|improve this answer
5

While I am not insinuating in any way that your lab does this in a malicious and self serving way, this reminds me of the notion of citation cartels

Its quite possible that this is an artifact of the research itself and its quite alright to self-cite (or group-cite) a few papers especially when one work builds upon the other. However, there must be more papers in the research community outside your own lab that must be citable? Surely, not 86.6% citations must be from within?

Speaking from my limited experience, if I were a reviewer and I saw papers with this kind of a pattern, I would certainly suggest the authors to include specific, relevant papers (which I have knowledge about) from outside their own lab in the literature review. I might of course be completely wrong and someone else with more experience in academia might have a different opinion.

|improve this answer
  • 1
    Aren't you confusing "citations (from others)" with "reference items (you put at the end of your paper)"? Either way, are you serious? I mean, would you write a report that says, "Cite more papers that are relevant to this result. I don't know which or where you can find though"? That's the most annoying kind of report. Wait. There's even worse. But your report should be based only on facts and the content of the manuscript. I agree 86% must be unusual. But "Whoa, whoa, that looks kinda fishy, you know what I'm sayin'?" isn't going to fly in science, at least without evidence, I don't think... – Yuichiro Fujiwara Jun 2 '13 at 2:07
  • No, I am not confusing it. References in one paper are citations to those specific papers. My grammar might be off though. Often, when going through the review process I get directives from reviewers which ask me to look at other specific papers which might be relevant etc. I don't think there is anything wrong in that and is quite a common practice, at least in my field. I would be a tad wary though of papers which have such high proportions of self citations or related citations and which (if I knew them) did not cite other works in the field. – Shion Jun 2 '13 at 2:36
  • Hmm? OP said one paper from their lab got 15 citations in total, 13 of which come from the same lab and 2 of which are from strangers, so to speak, right? Do referees in your field say your paper sucks and tell you to look at other relevant topics next time simply because your past paper got only citations from within your group? Without evaluating the merit by reading the content?? – Yuichiro Fujiwara Jun 2 '13 at 2:48
  • Also, even if OP means 13 "references" out of the 15 were their lab's papers, I still find it surprising that a referee would ask you to cite papers that they haven't read or even know. It may just be me, but to me it sounds so rude to question if the author did their homework when the referee can't give any concrete evidence. If the referee gives a concrete example, then it's either that the author's fault or that the referee's recommendation is wrong. Either way, the author can prove themselves right or revise the manuscript accordingly. – Yuichiro Fujiwara Jun 2 '13 at 2:55
  • 2
    Pure mathematics might be a different ballgame but when you work in nasty interdisciplinary work like I do, having a limited range of citations can be quite a drawback. As I mentioned previously, it is a common practice in HCI or information science for referees to recommend certain papers for citation in your article. e.g. Right now, I have 2 papers in publication and referees recommended 6 new and specific articles for citation. I should make it clear that what I meant was that referees often ask to cite specific papers and not generically say that you need to cite more papers. – Shion Jun 2 '13 at 3:20
4

If I saw a pattern of only or mostly citing work from your own team, it would make me suspect that the authors have poor awareness of related work. How likely is it that there is no other relevant work being done by anyone else in the literature? Nothing else that has any relevance whatsoever (techniques, methods, related results, inspiration)? That's awfully hard to believe. It would make me suspect that the authors are either ignorant about the field, or are doing a poor job of citing related work.

It's not clear to me what you meant by the 13 of 15 statistic. If you mean that they wrote a paper or dissertation where 13 of the 15 papers cited in the related work section is to people within their same group, yes, I would tend to look down on that. It isn't necessarily wrong, but it would certainly make me ask some pointed questions and make me suspect that there is something wrong. It is important to be a good scholar and to understand other work in your field.

Maybe you've heard the saying: "a week in the lab can save a day in the library". (This is not a recommendation!)

|improve this answer
  • 3
    I'd also have this concern if a substantial portion of the references in a paper are to papers of the own group. But this doesn't seem to be an issue here. – silvado Jun 4 '13 at 18:56
1

If your lab is the leader in a certain field, you would be aware of it. If your colleagues are pure geniuses, you also would be aware of it.

In both cases you won't be here to ask about it.

In others cases, extreme repetition is not a good sign.

(Keep in mind that little repetition is good as it shows interest in a field, but 13 in 15 is a red flag, a big bad red flag with a big bad skull on it)

|improve this answer
1

I think it really depends, is it justified or are you just citing each other to rise your charts? I think if they are justified, it is fine. Some labs really do have famous works and they just add on it. I think it would become the advisor's problem after a certain point, which would mean he has to renew his work. That said, it doesn't look great when people who have their citations come -only- from their previous work.

|improve this answer

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.