Hi Dear Ramesh

I think that self-citation is shown reflects the author's ability and power.

The arguments generally pulled in towards the self-citing a document is mainly for two reasons, one, a researcher has got every reason to cite his/her earlier works which forms the basis of ones present study, especially when the researcher is carrying forward his/her earlier works or continues to further his/her research findings in a chosen field. Besides, citing ones own work somewhere particularizes the research interest of a person and helps in bringing together the similar works. while as the negative aspect of the practice is h-index and Impact Factor (IF) which has somewhere become the parameter to judge the quality of a research work, which i fear warrants introspection.

Agreed with Pandita. In fact there are situations where self citation is main citation on which the current paper is based. The problem is not with self or other citation but is with indices like h-index. So there is a need to bring in indices which will overcome this problem (if possible).

Ramesh - in one form or another, this is a question that has been asked previously on RG. At the end of the day, self-citation is fine as long as it is not simply 'self-promotion' of ones own work to the exclusion of other seminal or current works - and that it doesn't bring in an inherent bias into the material presented. I often self-cite, although less so more recently. The reason for this is that my work was once the most visible out in the public domain, whereas more authors are more visible in recent times - so I feel more inclined to cite them now - especially if their work is current and builds upon my past work.

Recently I have published an article entitled "Effective Strategies for Increasing Citation Frequency" which is available online on http://ssrn.com/abstract=2344585 . You can find over 33 different ways for increasing the citations impact.

Ahhh - but Nader, as my last response, should it be less about 'how to increase self-citation' and more about 'when is it appropriate to self-cite?'

@Dean, Thanks for your comment. May be the following fact reply your question.

"Thomson Reuters resource known as Web of Science, considers self-citation to be acceptable up to a rate of 20%, anything over that is considered suspect” (Diana Epstein, Impact factor manipulation, The Journal of the European Medical Writers Association, Vol. 16, No. 3, 2007)."

We cite (self or else) when it is must for the build up of the paper. Whether a particular citation is redundant or not can be checked is not it? Citation is citation as long as it is relevant. but to give a cutoff like Nader pointed out as maximum 20% is some how not rational. As I said in my last post that there may be paper with much more than 20% perfectly honest and relevant self citation. So how to decide!!

@Subrata, I agree. "Citation is citation as long as it is relevant." however, is some discipline, even 5% is too high. The cutoff point for any discipline can be retried from Thomson Reuters JCR.

Hi Nader,

That sound seasonable to me. 20% is generous from ISI. I would split the difference and suggest that 10% is on the border of being acceptable - as long as it fits the criterial already discussed in this thread.

Refer to "MAXIMIZING THE IMPACTS OF YOUR. RESEARCH: A HANDBOOK FOR SOCIAL. SCIENTISTS. LSE Public Policy Group. (2011)", Self-citation rates across groups of disciplines is varied. Personally, I believed the acceptance level should be less than half of the attached table.

Thanks for the comments

Self citation may be encouraged only for citation purpose, but not for calculating 'H' index.

I'm with you Nader,

My discipline is the lowest 'medical and life sciences' - but i would still say half - and that is back to my original suggestion of around 10%.

@Dean, I'm with you too,

@ Dean, Recently i undertook study on citation analysis of oncology publication across continents for the period 2003-2012 and while evaluating self-citations in the field i made following observations.

Of the total citation 5249511 received by oncology publication at global level during the period of study of them (1502900, 28.62 %) are self cited, which also means more than one-fourth of total citations.

Of the total self citation received at global level, North America leads the table with a share percentage of (955458, 63.57%), followed by Europe & Asia with a share percentage of (354567, 23.59 %) & (169250, 11.26 %). Africa and South America have less than 1% self citations, while as Oceania has (16933, 1.12 %) self citations.

The self citations share percentage of the continents of the total citation received by publications at respective continental level, North America leads the table with a self citations share of 45.12 %, followed by Europe and Asia with their respective self citations share percentage of 16.02 % & 23.07 %.

Africa has a self citation share percentage of 9.25%, Oceania 13.60 %, South America 11.61 %. North America is the leading continent which has maximum number of average 9.28 self citations in each oncology publication, followed by Europe and Asia with 2.84 and 2.39 average self citations respectively. Oceania has 2.31, South America 1.59 and Africa 0.85 average self citations in their each oncology research publication, published during the period of study.

Self-citations prevent that the same information is published twice (e.g. details related to methods presented in another publication) and will happen more frequently when a study topic is explored by a single researcher/team.

Perhaps there are biology-based or culture-based underlying mechanisms influencing individual self-citation rates (e.g. personality profiles....)? The underlying mechanisms of self-citation rates would be an interesting topic to be (further) explored in human sciences/ human psychology....

@Pierangela, I have written (Citation Frequency and Ethical Issue, http://dx.doi.org/10.6084/m9.figshare.1027327) a letter to editor and raised the issue. I do believe that the author should follow the ethical issue rather than the evaluator.

informative discussion and exciting data for thought. Thanks to Nader and Ramesh

@Subrata You're welcome.

@Pierangela Thanks for your kind feedback.

The practice of self-citation is actually much larger than what mostly researchers across the globe perceive, even though it wont be perhaps inappropriate to say that all type and kind of citations are actually influenced by the element of self citation, which researchers have already rated as, author self-citations, journal self-citations, institution self-citations, regional self-citations, country self-citations. And who knows tomorrow we may talk of linguistic self-citations, subject self-citation, group self-citation and many more.

I think that self-citation is allowed only in one case, to avoid repeating information that is relevant to your article and published previously.

@Alexander, It is true for author self-citations. However, what about other types of self-citations such as, journal self-citations, institution self-citations, regional self-citations, country self-citations.

Well, it is another and more complex question. In my practice, I was faced with the first two. In this case, to my great regret, I have to compromise.

Citation is a way of giving a credit to the original source. If your work is the original source and you wish to extend your previous research findings or fill up a new gap grounded on your previous studies, self-citation will take place naturally. In my opinion, self-citation is permitted as long as the citation meets the research context meaningfully.

Hi Dear Ramesh

I think that self-citation is shown reflects the author's ability and power.

Self-citation should be useful and allowed when it can assist the author in avoiding repetition of his/her previously published research work. Moreover, published work that is either highly cited or published in a high quality publication then it is only reasonable to expect self-citation. However there are many cases of pure self-promotion which is not that ethical; but it isn't illegal

Agree with you both Alexander and Yurgos. The low self-citation in Africa should be examined wrt amount of published work on the continent.

Researchers may use self-citation to manipulate the h-index.

Self citation, like editorship of journals, may produce bias.

Within scholarly communication, there are practices and processes which draw criticism. Just think of the debate about peer review, for example. Criticism usually stems from something being done badly / poorly / not in the best interests of all concerned. Self-citation is useful, for example, when you want to reference either previous or related research you have done on your current topic. An issue arises when it is used "poorly", e.g. to "game the system" and manipulate the h- [or other] index. Perhaps the challenge lies with how indexes "value" self-citations rather than whether an author includes them ...

some times self citation is unavoidable, often forced to increase so called impact. At the end you or your paper is as good as the other (not self) say. It should not be tough to have indices after adjusting for self citation or may I say only keeping honest citations. Man by nature selfish ...other we would not have seen what is going on in around us in the world.

It depends on the case, sometime when the author highlights previous work or extended work that is fine :)

If your paper builds on a body of research that you have published over a period of years, it would be dishonest not to cite the earlier work. Self-citation is also a good way to keep the number of references down to a managable size.

I have reviewed articles, chapters, and books. It has never occurred to me to complain about the practice of self-citation. If I were to see a case that was blatantly abusive, I would say something, but I have not been in that position.

Scholars who are dedicated must look again at why we cite particular works in our research papers? The essence is to comb for rich ideas from those works that can enhance the beauty of the new research, giving it a strong theoretical backbone, deepening its essence and weave it with existing theories and concepts in the area already grounded in academic arena. This can be done by self-citations or citing other works if they would serve these important functions. However, if citations are done not for these purposes or intents but are done because of some index factors, then I am sorry. Such researchers driven by such index impacts and factors must revisit the fundamental reasons for  research works and their publications.

Thanks

Reasonable and ethical approach in self-citation is the essence of publishing integrity.

Self citation should continue if he or she has good amount work which can be referred.

Self citation is necessary, but must not be taken into account in any evaluation of the author's production and impact

When you work on a topic you established and work in a continuous progress, self citation is needed. It will help readers to see tho pattern of your progress. 

I do not mean any impact factor increase, just the need.

Sometimes your work might be ground breaking, so you would need to cite it in your continued research in developing the work from its foundation.

Best regards,

Debra

I doubt it matters much. As someone who has self published, citation itself is an interesting phenomenon. I t matters in academia, but not always outside it. One piece of research I did was good enough to have the general counsel of a large legal association contact me for advice, to be quoted in two news stories, to have the most prominent  national immigrant rights organization in the US "cite" my study several times in a legal complaint to a federal department of government, and I am not an attorney. However, it has not been "cited" precisely because it was not submitted to a journal, primarily because I doubt there is one that would allow the length and subject matter to be considered. If someone thinks that is less credible than an academic citation for this subject matter; please name the journal that has the expertise to accept it for Submission: Exclusion of Indigenous Language Speaking Immigrants in the US immigration System.

And I agree that if the author has contributed to the same body of work, self citing shows a consistent focus or a progression in scholarship. If one worries about how that does or does not raise citation numbers, I think they shouldn't be researchers. The purpose of scholarship is discovery and knowledge, not recognition by numbers.

Here's a blog entry that underscores the point made by Blake on July 20 (above):

http://blog.impactstory.org/four-great-reasons-to-stop-caring-so-much-about-the-h-index/

Though I wonder whether that blog entry genuinely describes real academic practice in terms of how researchers really do their _academic_ work. Surely most people know their topics pretty well, and if you read any article on a topic that you know, then you also _know_ how it fits into things and whether it is any good or not. No one needs to Google the author's h-index to aid that evaluation! The idea seems nonsensical. On the other hand, academics do care about _their_ own h-indexes and similar metrics because _that_ is how their employers evaluate them, and thus _that_ affects whether they can feed their families and keep a roof over their heads. Administrators want quick and easy ways of "measuring" the researchers they are charged to monitor, and since they really don't understand the research, the quick and easy ways they design are wildly divorced from the realities of the way knowledge is built. Researchers, of course, also judge each other ("Yeah, their book on that was great, though their article from last year was a little nuts, no?" Etc.), but it's more complex and nuanced: great perhaps for understanding the ins and outs of conversation and debate in the field, but not for jotting up statistics as though for pop charts or sports teams. ;)

In my discipline (archaeology) self-citation becomes more and more necessary, mainly due to the limitations some of the printing houses put on the (re)-use of images. Since we work very much with ancient materials and comparisons the illustrations are quite essential. Unfortunately many publishing houses do not allow the (re)-use without licensing fees. While the proper reference to the original publication is self-understood and good scientific practice, paying for the use of images (sometimes actually originally created by you) is not OK and practically impedes science!

Instead of repeating images (sometimes very useful side by side for comparison), now one has to make a (self)-reference and the reader is obliged to collect a stack of books around him/her for looking at the referenced pictures.

Thanks for that, Wesley. At best, the situation you describe in your second paragraph can be seen as the naive and/or lazy following of unfortunate trends. At worst, it is language and/or nationality-based discrimination. The sad part is that many Mexican institutions are buying into it, in part because they are chasing the carrot of international rankings.

I think sometimes self-citation cannot be avoided. If your research deals with the basics of a topic and develops then maybe no-one else has done it or you want to refer to some finds you published earlier and which can help to explain some feratures on which the actual paper wants to deal with. For some students or scientists who don´t have acces this might help to get an idea what else was published and contact the author to send the file.

Of course if someone is citing more than, say, 10 papers of his own this seems to be exaggerated ;-) I found some people are only citing themselves to be cited at all and to show that the work of others is worth less. This of course is the worst case, but thank goodness the exception.

Thanks Susanne - indeed, when only self-citation is practiced (or in an exagerated way compared to other citations) that is hardly to be taken seriously. I also know several such cases.

I would perhaps like to mention another thing - of course we all know that the statistical "number of citations" is not a good way to measure the importance of a study. Just as a negative possibility: if a very bad study is published it will be widely criticised - and consequently also widely cited in the critique. That of course raises the "citation index", but not at all the quality!

On the other hand, an innovative, ground-breaking study often draws criticism from colleagues working in a traditional paradigm that no longer holds up to confrontation with available evidence. Thus negative criticism is not necessarily a sign of poor research. Scientific research is not a popularity contest, and quality cannot be measured by the academic equivalent of "likes" (which citation indexes are coming dangerously close to).

Interesting comments, Nicholaus and David! I would add that even fully justified negative criticism can be useful. If a published study is bad because of poor choice of methods or incorrect / unjustified assumptions, then citing it as an example of what not to do can contribute to the field.

(good to see archaeology being represented in the answers to this question)

It is so ordinary to do self-citation and in many trusted academic sites they are presenting the two type of citation

But I encourage you to self cite your previous publication to improve your h-index and to increase your paper visibility. I agree with Dr. Nadir post.

Regards, Emad 

Many works are based on or additions to previous work. Self-citation is then necesssary.  The readers will judge whether it is valid or gratuitous.

I think you're correct, Roderick!

I believe that self-citation is a way of pointing people to previous published work you have done in the subject, that seems relevant for the article in question. It should be irrelevant whether that adds to any measurement system or not. 

We seem to be too worried now about evaluation systems, but as Wright-Carr and Boroffka mention, we should not concern ourselves with the citation index systems of assessing the papers, but rather base our decisions on the content of the papers.

Given word limits and all kinds of other constraints, it's really difficult to say anything new in a single paper; if I couldn't cross-reference to my other papers, I might as well just give up.

As for the obsession with impact and all of that... is that really the best way to judge the quality of someone's work?

Self citation without sufficient peer review should be questioned and should be monitored by the journals in which the research is published. I think it is reasonable to cite papers written by yourself and one or more authors in your own work. However, I do not think referencing papers where you are the lead and only author is valid, unless they have been passed review and published.

In my opinion, self citation is appropriate if it is fit, justified and relevant to the purpose of the study; and if that citation has been peer reviewed before being published. As an ethical attitude in the personal realm, the concern on the impact factor should be excluded.

IF YOU ARE THE ONLY WORKER IN A SMALL FIELD IN  PARTICULAR COUNTRY OR REGION THER IS NO OPTION THAN TO SELF CITE

Discretion is the key:  as my namesake implies (above) where self-citation is relevant, e.g. in a limited field where there are few workers, it is entirely apposite.  But sometimes writers include a whole stream of references that are barely relevant: such trumpet-blasting is immediately obvious.  

I agree, self tied work must be directly relevant to the subject of discussion

it is obvious to most readers if trumpeting is the game. i must admit I am not keen on 'anonymous' reviewrs asking for citations of often tangential works being cited. This usually identifies who they are and what their game is.

In Serbia it is required by the Ministry of Science, in order for them to accept works as "scientific". In practice, it means many people citing their own shabby research in order to support unscientific conclusions... but the Serbian ministry encourages it. (Naturally, when relevant to the work, it should be allowed.)

If we all stick to what´s called "good scientific practice" no problems arise from self-citation. For me it was often times very useful to learn more about a specific authors work, because he cited himself. But I do also see, that it is sometimes more about self-promotion than research. However, to ban self-citations can not be the way to avoid that. I would agree with several others here, that it has to fit properly and should relevant to the study.

Dear Ramesh,

The practice of self citation is definitely a good practice.But I have doubts whether it should be  made compulsory .Aleksandar Boskovic said that the Serbian ministry has made self citation compulsory for accepting scientific works.This means it has accepted self citation as a standard practice in scientific research. .Self citation shows  that the author is an active researcher who has previously done work in this field .No doubt , as Robert Armstrong Osborne ·rightly points out, self citation should be relevant to the work being pursued..I agree with Robert Armstrong Osborne that if you are the only worker in a small field then you have no option but to self cite.I , for  example, do research in student radicalism.Apart from doing Ph.D. , I have done two Major Research projects on student radicalism .I have also published monographs like "Students & Radical Social change "(2003) and "Change & continuity in Student Radicalism".But, in the twenty-first century,  I find very few works in this field.So, I have no option but to self cite.I think that self citation is definitely a legitimate academic practice.It should encouraged, not  discontinued.I thank you for bringing up this important issue.

Yours sincerely,

ANIRBAN BANERJEE

Professor of Sociology,

The University of Burdwan,

Golapbag, Buran 713104,

West Bengal,

INDIA

Why not have 2 citation indices - one with self citations and one without? Just need to get Google Scholar on side and then which is being used can be easily checked. 

Sounds good

I agree (as a worker in obscure regions) that self-citation is unavoidable. No doubt reviewers can step in for any egregrious self-promotion.  And it does help locate a body of work.  Peter Kershaw's solution really underlines the fact that  H indices (and the Research Gate ones) should have an error term stated.

I agree with the comments from Peter Kershaw and Geoffrey Hope (above), which are likely to reflect the views of most serious researchers. Also the role of reviewers is important. In my experience excessive self-citation most often arises in discursive or review papers and less often when research results are being presented.

Ok if reviewing. Ok if publication is building on or taking  further ones own research.

In many cases, self citation becomes a must. Suppose, a scientist notices an error in his/her already published paper. To correct the mistake the author will have to cite the paper in which the mistake has occurred. A scientist works on a particular area and publishes papers on that. When he publishes a new paper, he is to mention his/her other papers in the literature review along with papers published by other scientists. Many a time self citation is a necessity. We have done citation analysis of thousands of scientists nominated for various awards, and seen that generally self citations do not go beyond 10% of the total citations.

Any citation is only appropriate when it is 'relevant' to the article being written. If you work in a subfield which is populated by only a few researchers, then you will likely end up citing your own articles a lot. This is not a bad practice, just a result of the fact that you are a major contributor in that area. Excessive and unnecessary self-citation (citing your own paper when some other paper would be more appropriate) is explicitly bad research practice.

I am personally in favor to allow it. There are three reason for it-

1) If author is extending his/her own work, then definitely self citation will be there.

2) There may be some use of particular technique of author's previous work or new work may be related to the previous work, then author should put it (previous work) in the related work section. 

3) One last reason, everyone want to show his/her own performance and want reward on it. Therefore citation is one of the criteria to do it. 

I support self-citation in that, in addition positives advanced by other discussants above, self-citation helps readers to appreciate context-specific situations of a subject matter under discussion.

However, where it becomes superfluous and clear that the author is using it MAINLY to enhance his visibility and impact, editors and other appropriately approved sensors should step in and regulate the practice but not to abolish it.  

One cannot avoid it, e.g. if one uses a research technique  already described in an earlier paper one would not repeat the whole description.

I think there are already good comments, e.g. Bernd-Ulrich Meyburg, Md. Sohail, Bimal Kanti Sen.

Self citation may be allowed. This increases the author visibility. But should not be a situation where it looks like as the author is presenting Inaugural Lecture

Scholars normally cite their works because of the following reasons:

1. They are ignorant of the importance of citation, mostly as it relates to the ranking of universities in the world by the Times Higher Education.

2. To avoid self plagiarism

3. They want to update their previous studies.

A reviewer once pointed out that I didn't cite an important paper in my manuscript - it was my paper. I had already self-cited one of my papers and thought it was not appropriate to cite another one. So, although my cultural upbringing consider self-cited as tooting one's own horn, sometimes it is necessary, especially if my work contributes significantly in the development of my argument.

Edmond Sanganyando has just presented the best solution to the problem of self citation.  I did not know that till got to know the significance of citation; since then, I have stopped citing myself. I leave gaps in knowledge in my previous studies for other scholars to identify and fill.

Self citation is allowed because it shows the strength of the scholar in previous works. Therefore, I do not see anything wrong in it.  Tt is a case of if you do not blow your trumpet, then who will do it for you.?

I believed that self citation is not scientific work because previous your work is important just for you not for other society or other researcher. 

Well, the idea of Prof. Gberevbie  is equally good, but such a view could be applied to  Third World countries where devilish scholars, irrespective of religion, are most interested in suppressing potential materials. If such materials fail to blow their own trumpets, certainly, it will perish. However, such persons must exercise caution on self-citation because enemies can still use it against them, since citation is one of the criterion for ranking universities in the world. . 

It should be allowed. However there is need for guidelines on when to self cite and when not to self cite. Sometimes your previous research is an integral part of an argument in your paper, in that case self citation is ethically allowed. Sometimes there is no other relevant data or information except your own publication. On the contrary, you can not cite your work just for the sake of citing even when it is clear it is irrelevant. To me, it is all about ethics just like the process of carrying out the research itself.

I think the scenarios may differ, as much as self-citing has its scholarly objections, there are situations one need previous personal work and to serve as the leverage for continuity of a research activity.  Especially, in the case of doctoral candidates presenting an article based thesis - you cannot afford to have a disconnected articles. Ethically, self-citing needs to be done on the grounds of relevance and promoting grounds for better comprehension of an on-going or subsequent related works, and not just for the sake of it. 

In my opinion, self-citation is dependent on quality and relevance.  

• Self-citations are both appropriate and needed when they provide essential / foundational / germane information to presented topics. 
• Conversely, when self-citations deliver extraneous / marginal / superfluous information to presented topics, they become unfortunate and very obvious attempts at author self-aggrandizing. 

The latter should be discouraged / eliminated in modern published literature.

One needs to cite himself if a portion of your earlier work is needed in the current one to reinforce a fact or an observation. If you fail to do so,you can be accused of self plagiarism.Citing oneself shows your strength of knowledge in a given area; It enables readers to know your contribution to knowledge  in an area of study or research endeavour

The current work is an extension of the author previous work. This point need to be addressed by the self citation frequently. Or I have looked to this problem from a specific angle in my previous work and I want to look to it again from other side of view.

If your work is good enough for researchers to cite, I see no reason why you can't cite your own work if it is relevant to your present work. What should be discouraged is self citation not relevant to the present research or any other non relevant references.

If we beleave in continuation of researcher work, which most of the times leads to innovation, then we must appreciate self citation.

Dear Ramesh;

I have some research, development, demonstration, and deployment projects. These projects are long term projects. Some of them are life long studies to me. 

I have to cite my own publications to explain the improvements and present new findings to our World, otherwise it is meaningless and useless

As far as I find according to my literature review, I am the only one or one of the few researchers on the World who works on that subjects. 

There is not any other option for me. Otherwise I can not present my findings properly. 

I only work on a few projects related with a few topics. 

Please see my projects and my research publications. I think they are very good examples for what I try to explain. 

I hope one day, I can present best %100 renewable power global grid and best power plants for our World. 

Have a nice day

The criteria for citing one's own papers are the same as for any other references. Any references necessary and relevant to the present paper should be cited, regardless of who is the author.

Any references irrelevant to the present paper should not be cited, regardless of who is the author.

This has nothing to do with whether the cited paper is good or bad. In fact, it is sometimes necessary to cite a bad paper in order to point to its faults and reject its conclusions. It is bad science practice to just "let go" and fail to mention a paper just because we know that it is of poor scientific quality and/or proposes the wrong conclusions.

Some journals already ask the reviewers if the author does not abuse the right for self citations. It is OK, but the more restrictive politics would be wrong, as my colleagues have explained in the previous comments. However, sometimes I come across the papers where self-citations constitute more than half of citation list, which should be considered wrong or at least suspicious. Then I indicate it in my review.

I find no harm from self citation unless there was an abuse manner.

Self citation should be encouraged provided it is contained within the limit of normal.

During my time as a researcher in the rubber industry I published a number of review papers for different orgaisations. I found numerous researchers who were obviously misusing citations. This seemed to be strongest in the field of statistical applications, That said people in engineering operations and at University Institutes were no slouches. However if work is ongoing then reference to past work is essential and this leads to self citation. If you do a lot you are going to cite yourself often.