A friend of mine, who has been the editor of the main journal in my field has the opinion that a person who is coauthor in a paper, should be able to defend and understand all the main aspects of the paper. That means that the person has been deeply involved in the main aspects of the project. I think that this is a good definition of whom should be considered as coauthor.

There are many resources on this, most of which suggest similar things. Here is one resources

http://www.phdontrack.net/share-and-publish/co-authorship/

Simply securing grant money does not justify co-authorship by itself. There needs to be direct contribution to the final product. The same goes for individuals who provide logistical support. They contributed to the research but not the final product. I think it's important to consider that a publication is an advertisement of scholarship and not the scholarship (aka research) itself. For individuals contributing to the latter, it should be a great honor to be acknowledged in any resulting publication.

Who can be a co-author? A scientific article should have a message, an intellectual idea. Co-author can be he/she who does contribute to the conceptual essence of the future article’s message. Others are mere cronies, bosses or pseudo-intellectual coordinators.

Andras, so strong opposite? Even to aperson, who just get money for your research? Then why he or she should bother about your finances?

I disagree with Andras that "Others are mere cronies, bosses or pseudo-intellectual coordinators." There often is nothing "mere" about their contributions, as they provide the infrastructure for scientific research. They should get an unbelievable amount of credit for such a contribution, but this should occur in the acknowledgements of the paper.

After all, it is not common practice to provide government or scientific agencies co-authorship for awarding a grant to a researcher. They provided the infrastructure, and they should be put high on a pedestal for doing so. But it is the researcher who did the work and interpreted the results (and possibly made recommendations for new researcher). Even the director of a laboratory should not automatically be given authorship simply for providing four-walls and a computer. They need to make another contribution to the intellectual merit of the actual work. Otherwise, we should be giving authorship to the printing company or the computer company as well.

I think this is an increasingly tough question especially in large scale genetic/ genomic/ biomedical studies, in which there frequently may be 100+ authors. I have heard from some authors on such papers that a contribution may have been to write a key but very small piece of code, perhaps one afternoon's work. Do you think there is a minimum time/ effort requirement for authorship? I have also heard cases where a co-author did not even know they were on a paper until after it is published! Also what about contributing a key piece of material or key re-agents, or being the main provider of funding?

I'm not a good enough coder to do anything with genetics, but I have contributed code to many many many people. This is often code I've written in my spare time or for other research. Sometimes, I might give a few hours or a couple of days. Most of the time, I've spent a fair amount of time working with them so they know the logic behind the code and how to reproduce and adapt it. But my policy is not to request authorship for such a thing. It's just part of the academic game. In most cases, I've made additional contributions that have elicited invitations for co-authorship, and I think this is great and justified (and I accept).

It is indeed tricky with large genomics datasets. But, again, I believe there is a clear distinction between research and the advertisement of research (aka a publication). The advertisement of research takes more than vocational skills; it takes analytical skills, and/or writing skills, and/or skills for interpretation, and/or skills in background research, and/or skills in coming up with the original idea, etc. My over-use of "and/or" is simply to say that publications are multi-faceted. We should certainly thank (profusely) any contributor of a piece of code or a someone who did lab work for an afternoon, but we should acknowledgement them as contributors only. They did not contribute to the meat and potatoes of the published work. Maybe a metric for co-authorship should be:

- Did you author the paper (actually write it)?

- Did you author the original idea (write it in your head)?

- Did you author a grant to get money?

- Did you author the entire analysis workflow (from data collection, management, to final product...including getting little bits of help from other writers of a "single line of code")?

- Did you author the tool used to conduct this particular analysis (this is tricky, but you could change this to mean..."did you author the technique used in this paper and did you actually use it on this particular data?")

Then, you need to do two of the above. This doesn't differ from other lists out there, except that it puts the list in terms of "authorship" rather than contribution.

Good discussion.

All those who played a meaningful role in executive the project and write up of the article

A friend of mine, who has been the editor of the main journal in my field has the opinion that a person who is coauthor in a paper, should be able to defend and understand all the main aspects of the paper. That means that the person has been deeply involved in the main aspects of the project. I think that this is a good definition of whom should be considered as coauthor.

I read an interesting comment on this by Jerry Coyne on his blog:

http://whyevolutionistrue.wordpress.com/2012/05/09/mohamed-noor-profiled-in-the-scientist-and-a-note-on-authorship/

@Joachim. Excellent link. I think everyone should read this. This is my belief also. It's good that there are people out there who still believe this way. Thanks.

In a small research environment it is very possible to have a project that belongs to one or two individuals. In a larger research environment there is a great deal of collaboration going on. While the design of an experiment may be done by one person, the execution will be done by another. You can also add to this that sometimes the methods must evolve with the research and that is just the analysis side of things.

Either way, when it comes to write up the experiment for publication, you, as main author, should try and engage as many of the people who helped you on the project as possible. Did you write up the statistical methods right, or how the samples were collected, especially if they were from another site? In addition, engaging the people who others may write of as "technicians", will make them more engaged in the work you all are doing. Some of them may not want the added responsibility of revising papers, but they will appreciate being treated as more than just support staff.

Finally, please remember, there is also an Acknowledgements section of your paper where you can thank all those who helped you, but did minimal effort in the final product. There may be someone who helped with w great insight for your discussion, but otherwise had little to do with the paper. If they do not make the cut as author, but you feel you could not have done the research without them, mention them here.

I have not read all the answers before me, but I suggest reading the Vancouver protocole :

http://www.authorder.com/index.php?option=com_content&view=article&id=28&Itemid=47

"The Vancouver Protocol was first described by the International Committee of Medical Journal Editors and is now applied across all disciplines in the world's top universities. "

Linas,

I had two statements. The first one was the more important. The second one said: the persons who do not contribute to the real message of a scientific work are cronies, bosses etc. Interestingly, nobody concentrated to the first one which was the carrier of the scientific merit: the idea. You stressed the money which may play a role in the implementation of an article but it has nothing to do with the message. I think one should make a difference between the scientific contribution and the economic contribution. A sponsor or a Maecenas should not be an author. By the way, I propose to read an article on the commercialization of the science and research. There are a lot of them.

Patrick,

I think if somebody is a boss and makes his/her own task should not be a co-author. I note some bosses do not do much for providing a good infrastructure. Certainly, it depends on a lot of things. Naturally, general declarations can be understood only if one characterises the circumstances. However, I think in many cases some took advantage of co-authorship and this becomes an easy way to get to publications.

@Linas. I completely agree with the comments you addressed to me.

Contributions in making the project successful one entitles you to be a co-author.

All those who are active contributor in a research project may be included as authors/co-authors.

Anyone who made significant contributions to the research should be a co-author. The best way to decide is to ask the questions: (1) Could this project have been completed in the way that it was without this person? (2) Could this person stand before a group and defend the SECTION of the paper/project they contributed to?

@Jeanie, fully agree - if someone is co-author, he/she should be able defend his contribution.

But what about providing material (samples, data, interviews etc.) - should this be credited for co-authorship?

Dear Jeanie,

What you have written is almost perfect. I have only a question: What will defend the person who gave the money or supported “not scientifically” the major author?

Linas,

I repeat: What is the difference between the scientific contribution and the economic contribution in a scientific paper? Was the emir of Bukhara a co-author of Avicenna (Ibn Sina)?

@Andras, I am not talking about mecenats. My point of view was about the leadership - to get grants, in many countries you need to have at least one person in the team, able to defend position and be responsible for the final result when grant is obtained. Say, younger colleagues have insufficient capacity for obtaining grant (noit enough publications, for example).

Here I have in mind, that leader will work himself in the field, process data, give ideas or make other type of contribution in addition to getting support.

Then, why he should do this support, if he did not qualify to the paper (according to your point of view).

So, I think you misunderstand me and Jeanie - we are not talking about a PERSON, who gives money, just about a person who BRINGS money.

Thanks, Linas.

You were now clear. What you have written is understandable and acceptable. However, I mention, Jeanie has not written about financial matters.

Andras, Jeanie's statement was "(1) Could this project have been completed in the way that it was without this person?" - if someone secured funds, it is nothing?

Linas,

We can still wait a bit. Also Jeanie is able to answer.

So, based on this discussion, is there any consensus on what constitutes a co-author?

Yes, I disagree with Jeanie's wording of what it means to make a contribution (Re: financial contributions). I imagine that most, if not all, scientists don't think that authorship credit should be ever purchased. Of course, many lab directors are not only providing an infrastructure grant but also providing feedback during the course of the project, lab space, guidance during analysis, and help with writing. If someone gives money to the researchers--and this is their only contribution--, they should not be granted co-authorship. They should be credited heavily in other manners.

Again, I think it would be good to devise a specific list (perhaps building upon some of the lists in the linked resources above) so that this could be of benefit to future readers.

All those who have contributed significantly should be the co-authors, irrespective of the work performed, may it be experimental, literature review, manuscript editing, formatting, submission or handling throughout the review process. If the corresponding author feels that the contribution of a person is such that it would have an impact on the quality of the paper, the concerned person should be included as a co-author.

Alok's list brings up an issue that I think is important. Everyone seems to have a good intuition about what justifies co-authorship, but it is often very difficult to actually create a clear list of adequate contributions. For instance, preparation of the manuscript sounds like a good contribution, but I am concerned by Alok's wording that suggests that a someone should receive co-authorship for simply submitting a manuscript. I'm sure this is not what Alok meant, but I think it's important we all get very clear wording for what constitutes contribution.

@ J. Patrick Kalley.. Yes You are right ... submitting the manuscript was not what I meant. It also included manuscript writing, editing, submitting and handling the manuscript throughout the publication process. I believe this is a significant contribution as far as manuscript co-authorship is concerned.

Thanks and Regards

@Alok. Thanks for the clarification. We are definitely in agreement.

Alok and Patrick,

All these general formulations are all right. However, I think one can find the essence often in the details.

Voila: one of my PhD students gave me her manuscript in order to correct it and also to calculate the statistics. I was amazed because my name was indicated as a co-author. I asked her why? She answered: this is a custom. But it is not my custom because to help a PhD student is my task.

Another example: A boss of a department said one day to one of his colleagues (a subordinate to him): You have here two manuscripts to submit. Now, it is time to decide in which you will be the first author. The young man said to me that the bottom dropped out of his world.

Thus, it would be really great to prepare a list as Patrick proposed.

Fully agree, that second example is awful, and a list may help.

But the firs one is OK. If your student is not able to do statistics, this is NOT the task of supervisor, and co-authorship may be credited. See discussion about statistician's role, Rgate.

@Andras. I agree that the second example is terrible. Maybe it is also time to reprimand or fire that professor for a type of plagiarism (using someone else's words and claiming credit for those words.)

@Patrick, I vote up your proposal!

I totally agree with Ernesto. All the co-author should be able to defend and understand all the main aspects of the paper !!

What Ernesto wrote and which was repeatedly confirmed is a logical and totally acceptable demand versus co-authors. However, this cannot be controlled and if it would be a general and controllable rule many co-authors would get in a dangerous situation.

I am glad that so many like the answer, which is not mine, the concept belongs to a friend of mine. Unfortunately, that is the idal world, and so many times we have, we must, or we are forced to include authors that are not able to defend and understand the main aspects of an article, and there is when politics and power games come in the academic world. Unfortunately, many times those who has the power of blocking your research in the near future, play subtile (and so many times no so subtile) games to make their names appear in a publication. And then, there is that other horrible aspect of publishing, which is the order or authors... I have had so many problems in the past with coworkers who are not happy with the place their names appear in the publication, and sometimes those problems lead to a breakup in colaborations. I guess is better to breakup with someone who fights if their name is in the second or the fifth position that keep working with them, but meanwhile it consumes so much energy and effort.

Ernesto did bring up an interesting point about authorship order, which I think is relevant to the current discussion. As I don't believe in the advice that the lab director should always go last in authorship order, I just created a simple index to determine authorship order. (Note: This is probably one of the most stupid "index" measures every--since it's based on that wonderful mathematical operation called addition. And this is certainly in no way novel. It's just a fun way to resolve conflict in a horribly simple manner.)

Co-authors can usually agree to the specific portions to which they adequately contributed (data analysis, providing all the research supplies and helping with analysis, writing the paper). So, you have a list of contributions (like we've already discussed) that can be very long:

- Conceived of the idea

- Secured funding for the research

- Provided a method or supplies

- Designed analysis (different than actually crunching the numbers)

- Collected data

- Analyzed data

- Wrote the paper (any part > 1 paragraph)

Give 1 point to each co-author for each of the above contribution, and then calculate their total contribution. For the lead-author in this example, total=6-7. If total=1, they don't get to be on the paper. (This last point may be amended somewhat, as the person who writes the paper should definitely be list an an author.) I calculated this for my own papers and found that it worked perfectly for authorship order and credit. (Of course, I had a small sample size...heh heh.)

Anyway, this is just one solution. Likely, there are problems with it. My only purpose in sharing this was to say that there may be a really simple framework to decide authorship credit and order. I mean, if math is involved, it must be right...right? right?

Dear All,

The list of Patrick is interesting. It is really simple but this could be an advance of it. However, the listed activities theirselves may be very various and different, thus the 1 point an activity cannot be proportional or just. Some examples:

- collecting (experimental or ecologic) data may be the hard work of some years

- designing analysis in most cases and at certain knowledge level cannot be an enormous work

- securing funding is generally not a scientific or highly intellectual activity

I stress that collecting data may involve also the manipulation of literature data as well as the experimental data. The list itself seems to be practical, its major advance is its existence.

Although, I would like to mention that practically the most dominant author decides on the co-authors and their order. Last question: who is the most dominant author?

András, please clarify what do you mean by the most dominant author.

Ernesto,

This is simple. He/she who has power or influence in this question. However, answering with a question a question is a careful answer. I think, regarding the dominant authors, there must be quite a great variability and diversity. Thus, general discussions without examples are not easy to understand. As far as I could see your last comment, you must have not only pleasant experiences...

Many thanks for the accent mark. You has been the first one. Muchas gracias again.

That's easy, my keyboard is in spanish!

Regarding the "dominant author"... that is a very, very difficult question to answer, but I might said that the dominant author is the one coordinating the effort, the head of the group, the corresponding author. But that line becomes difficult to draw when someone is getting independence in a group.

Sometimes I think that the papers should be anonimous, and the only one knowing who is writing what, should be our institutions, in order to justify our salaries... but all the knowledge should be shared in an anonimous way in order to avoid biases related to the names of the authors...

@András (with accent and apologies)...I agree 100% with your previous comments concerning the weight. In no way did I mean that these tasks should be weighted equally. What about the following solution to "The List"? If one task comprises of more responsibility or effort, it can be split up into more parts (in order to maintain the simplicity of 1-point/contribution framework, rather than applying a weight to the contribution). This could be:

- 1-point for each 2 paragraphs contributed to the writing

- 1-point for collecting more than half of field data (or for every 20 hours of work, depending on project, or something like that)

- 1-point for securing +50% of total project funds

Of course, dividing up the large tasks might then require a higher minimum number of points for being a co-author. This could be the responsibility of the lead-author, who could also set limits (or ranges) for the number of co-authors allowed for this. In any case, I think the system would work for determining order of authors based strictly on level of contribution.

Another advantage is that everyone entering into a collaboration knows the minimum contribution that one must make in order to be a co-author. (On that note, it could also be a way of motivating undergraduates to work hard for their place...I can see this being theoretically a good idea but practically terrible in the case of highly competitive undergraduates...sorry to crack on you crazy pre-med students.)

Anyway, I know that "The List" needs (1) a better formulation and (2) a better name, but I now want to adopt this revised list for all future projects.

Ernesto,

I see you have understood perfectly the complexity of this situation. However, I am afraid good dominant authors do exist only under very democratic, open and fair circumstances. As to the Spanish keyboard, I think the accent is “merely” the normal cortesía which I respected always.

Patrick,

The accent has no importance. In case of Ernesto’s address the style of J.L. Borges told me: “Buenos días, muchacho!” And the muchacho liked it.

It seems to me that your construction of “co-author instructions” becomes continually better and better. The maturity of such an instruction packet could be achieved with using of some actual data and situations. Nevertheless, I am anxious that this packet would not be favoured by many dominant authors.

Hi.. This is a very interesting topic.

I would like to know what you think if say someone demand to be co-author just because he/she is the head of department or simply because they allowed the researcher to conduct their research in his/her lab? And they have no hand in the researchers work.

Manorenjitha, someone who demands to be a co-author for the reasons you are pointing, has no ethics.

Hello Manorenjitha. You have pointed out the perfect example of ethics and conflict of interest. One wants to improve his/her IF and he/she is the boss then what? Do we have any chances to justify and at the same time protect our selves from consequences! Ernesto, we all know that it is unethical but what would be the possible solution to this? May be a straight forward rule in respective institute can be established that if you have significant contribution in the said protocol then only ask for authorship.

Manorenjitha and All,

It would be useful to assess how often this behaviour occurs?

Only with good examples can we get nearer to the multiplicity of our uncomfortable reality. And still one thing to mention: some courage is needed to write on such subjects. Thus, my respects to the “warriors” of this issue.

I am pretty sure, that such cases occur, and this is criminal, not un-ethics!

@Manorenjitha et al. Good questions raised. Unfortunately, I think young scientists will never know how to deal with these issues without having first-hand experience. Of course, that's obviously not the kind of experience that anyone wishes on their worst enemy (a debatable point, perhaps). A couple of thoughts, without belaboring the overall point:

[1] If the director has not only contributed his lab but also the unique (and I emphasize "unique") resources of the lab, then the request for authorship might be valid. This all should be decided before the bulk of the work is done. I think we've all seen cases where such disputes have been the fault of both directors and the ones who did most of the work and could have been easily resolved with some better up-front communication.

[2] Document everything, unless you really really trust the person. I would trust my life with most of the professors and collaborators I work with. But there is a reason to archive your emails and also ask for email summaries about conversations. This protects both parties.

[3] If there is documented unethical behavior, then it's time to cut any relationship with that director/supervisor. After all, it's not worth it to have your name (if you've been ethical) associated with that person. This is not a trivial decision and WILL affect your entire academic life. An important point here is not to talk widely about it to everyone and their uncle, except in cases of criminal action. If it's simply borderline unethical behavior, then I would recommend only talking about it if asked. I feel strongly about these points and here are some examples:

- A friend of mine was less than a year from earning his Ph.D. when, due to unethical behavior by his director, the student quit. He was well qualified (with papers in Science and Nature during his Ph.D.), finished his Ph.D. at another institution in record time, and is now a professor at one of the top four institutions in North America. And many people know the story and the reputations of both individuals (guess which one is better).

- I myself ended up with no letters of recommendation from my undergraduate lab to get into my Ph.D. program, due to similar issues (and realistically because I had a rather brash response). (Just for the record: If these individuals are on ResearchGate and wish to raise this point again, I'll gladly take you to the mats.) All I needed to nullify them was a single letter from a U.S. National Academy of Science member. As an ornithologist, it's all about making sure you don't have all the eggs in one basket/nest.

In sum, I think there are professors out there who overestimate the importance of their letters of recommendation for young students. In my case, their letters would have been discarded outright. My unsolicited advice, for what it's worth, is to have good convictions but also be aware of what you're doing wrong. Maybe that means that you should have documented things better; maybe that means you were, in fact, behaving poorly or stupidly; but maybe that means your professor did something highly unethical. Make a decision to confront the director, explain your view, and attempt a compromise (ask him/her to proof the paper and contribute a paragraph). This could solve the problem altogether.

If all else fails, quit. Quit loudly. But, I stress again, you must be prepared to take the hits. They will come, and they will also come loudly.

But sometimes it's worth it.

Some years ago a scientist offered me to be one of the co-authors in his planned article. I answered him thanks but regarding I did not worked with the manuscript and the research work, I could not accept it. However, I proposed him somebody else for this useful business. Later, I perceived that except of the real author (the trader?) in all probability each of the co-authors must have had as much contribution and merit in the authority as me. Another example: I was asked to identify some insects (several hundred individuals) and the person who only mediated the samples and had no idea on the taxa in question became one of the co-authors. Is false/bartered authority like an iceberg below the sea level?

So it seems, that data collection is not the co-authorship guarant. Well, then the most popular answer should be "please, come and collect samples by yourselves"? The same about identification. If there are five mice easily identifiable, then OK, but if you are asked to identofy several hundred specimens? Why someone must do this? Even simpliest sampling sometimes include heavy logistic (I tried to get wolf samoles from Russia, for example).

I am a specialist of the Chrysoperla carnea species complex. It is a bit difficult to indentify. In this case data collection was not a challenge.

For being accorate I have identified 613 lacewing individuals in this job.

maximum number of co-authors I have seen ,was 100, in the paper in Nature, of the moderate length. So what, anything wrong with it?

The Vancouver groups made these "rules of publications", which most journals state must be followed. They have just been updated:

The ICMJE recommends that authorship be based on the following 4 criteria:

1) Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; AND

2) Drafting the work or revising it critically for important intellectual content; AND

3) Final approval of the version to be published; AND

4) Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

If people are not involved in ALL of the above they should not be authors.

http://www.icmje.org

Dear Lars,

These are very logical and valuable recommandations. I have visited immediately the home page.

I like the ICMJE list, but I think the list still contains many vague terms that lead to the kinds of problems we have discussed in this thread. For instance:

- What constitutes "substantial contributions"?

- What constitutes a critical revision?

- What constitutes "important intellectual content"?

Patrick,

Particulars are always uncomfortable and make new troubles.

Troublesome specifics are better for progress than well-behaved vagueries.

I agree with you.

Let me remind the last part of the question: "What about cases when co-authorship was offered, but then "forgotten"? I undersand, this is simply un-ethical.

There can be various examples:

1. Co-authorship was offered because of substantial contribution in a given work.

2. It was offered if he who made the offer wants to be a co-author in the co-author’s article. This is a simple barter business.

3. It was offered for the boss. This will be useful in the long term.

And there are many interesting and advantageous constructions.

Which one is ethical?

4. it was offered to get materials from other country. Otherwise, collection is hardly possible.

1 in your example is OK

2 I do not like it, but it is acceptable

3 this is unethical

Some remarks:

Example 2/a. Roles: g1 = genius1, the boss of department a; g2 = genius2 , his wife working at department b; b = boss of department b; p = PhD student belonging officially to department a, but doing his work at the lab of department b; g3, g4, etc... friends of g1 and g2 in another country. P works and writes articles and the others are co-authors. Certainly, g2 is co-author in all the articles prepared in department a.

Example 2/b. the technique: 15 scientists write each an article, they offer co-authorship to one another, each of them will have 15 articles.

Brave new world?

your 2/b is a rule, I will not tell you the names, but meny of them are on the Researchgate :(

I am strictly talking about scientific publications. I only have experience in the Electrical/Computer Engineering field. I don't know how applicable it is to the other fields ,,, Here is our formulation:

--- FIRST AUTHOR: The person solely responsible for the submission, tracking of the submission, main results, a huge portion of the writing ... Something like, the person that does 60-70% of the work ...

--- SECOND AUTHOR: A backup to the first author. The person that does 20-30% of the work. First and the second authors gather data, do experiments ... etc ...

--- LAST AUTHOR : Senior author that oversees everything, advises the primary/secondary authors ... Helps with the flow of the article and/or deciding the DESTINY of the article ... That is, figuring out, what the story is ... or, if the PLUG SHOULD BE PULLED. or, what will not work ...

--- MIDDLE CO-AUTHORS (3 to n-1) : These folks contribute SOMETHING. Writing, data collection, etc ...

Typically, the idea comes from the adviser (senior author), or the students (this is very impressive if it happens) ... We make the students #1 and #2 authors, and the adviser becomes the last author, and the other contributors become #3 ... #n-1.

I have seen very different practices from other colleagues in different fields though ...

in biological field, I think, 1st author is the main one; the rest is in alphabetical order to show, that they are equal.

Idea about the last author is good, I will think about this.

Tolga,

Who can control and verify the implementation of your points? When I was a student my supervisor told me: In research honesty is the most important thing because nobody can control circumstances of your results and the way you got them.

Dear Prof. Linas Balciauskas, A coauthor is that person who partially helped directly in getting the objectivity of the paper.

@Andras, I think what you are mentioning has a lot to do with the ACCURACY (and, REPEATABILITY) of the results. I can only imagine how difficult this must be in the medical field ! But, in my Electrical and Computer Engineering (ECE) field, results are a little bit easier to make REPEATABLE. In general (according to my formulation), PRIMARY author is always responsible for getting accurate results in the first place. However, the SENIOR author is always the WAIT A MINUTE, THIS DOESN'T SOUND RIGHT person !

In my experience , it takes a while for the students to develop the COMMON SENSE to make WAIT A MINUTE, THIS DOESN'T SOUND RIGHT decisions ... It might very well take 2, 3, 4 papers to develop this SENSE. In the meantime, the senior author (typically, the adviser) makes sure that, the #1 and #2 authors (primarily responsible for the RESULTS) didn't make a technical or logical mistake. This generally comes from wrong assumptions. I will give you one example. It will be again from the Circuits field, but, it should hopefully be a good representative example for other fields too:

*** For one paper, we were redesigning a DC-DC converter and the students kept getting excellent results. I was always suspicious from day 1, since they sounded too good to be true. I was the LAST (senior) AUTHOR in this submission, and the #1 and #2 students were responsible primarily for the results. It was going to be too difficult to build a circuit, so, it was a simulation-only submission.

*** On day 20, when we kept getting the same excellent results, although I had made the students try 10 things by this point. I started storming the #1 and #2 authors (students) with questions ... Turns out, they put such a small LOAD at the output that (it had only 10 mA), the DC-DC converter never had any strain and worked flawlessly.

*** When we increased the load to 2A, things started looking NORMAL (i.e., imperfect). From that point on, I kept checking the load every step of the way. So did the students ! The students had developed a very good sense for it, and almost never made that mistake again ... They always added a good load.

*** This is why I have the SENIOR AUTHOR in my formulation, who does these SANITY CHECKs. Ideally, this should be the person that doesn't have to write too much, so, (s)he is not bogged down with details, and can find time for these major issues.

*** In my experience, students almost always do a good job with the details, but, they have a tendency to miss the big picture a lot more often !

Did this answer your question ?

Tolga,

Controlling and verifying ACCURACY (and, REPEATABILITY) is a task of reviewers, too. However, I note this is a very difficult job mainly in quickly developing fields.

I meant nobody can control the level and the truth of a contribution of co-authors in a paper. Only the authors themselves.

What you wrote on general rules and tasks of authorship is fine. I am afraid your example cannot be applied for subjects when studies of several years are shown.

Interestingly, merely “small fishes” discuss this subject. Where are the “big fishes”?

Andras, whom you call "big fish"? According number of papers or according position somewhere?

ah, you judge from RG score? Don't you think, big fish have no time for score building in RG?

Linas,

According to both positions and impact points. I agree a big fish do not waste time in RG. Perhaps, his secretary...

Well, I already told about 2/b... as an editor, I sometimes get caught co-authors, who hardly know paper is about. Though, I never try to know, who are they, what position do they held. Or paper is accepted, or it is rejected. In both cases, co-authorship is not contentious.

Thus, if a person, who really wrote the paper, cannot stand for himself - no one else will help him.

Logically the first author is who supervised the work, written the paper and followed all the submission process, he/she could also act as the senior if this scheme of decision is not part of the research framework of the lab. 2nd and 3rd authors must be persons totally involved in the research paper and must know its content and logic, they could be students if they are not able to perform accepted papers yet, or people who are implicated in experimental data interpretations, programming, or modeling, or any theoretical contribution in multidisciplinary research projects. If the senior is not the writer he/she must be 2nd author since all the research work was under his/her control and decision, or last author if he/she have only read the paper and directed the publication choice

@DavidThomas...Example 1 this is a nice concise list. My own list (see a post above) seeks to make co-authorship decisions more mathematically simple, but I like how succinct this list is. The second example certainly suffers from ambiguously defined "contributions," but it also may be sufficient--and maybe perfect--for small research teams comprising of co-authors who trust each other. On that note, much of the discussion in this thread focuses on how to establish guidelines for co-authorship in order to prevent in-fighting, purposeful omission of contributors, or false claim to work. Maybe our goal should be one of natural selection: We act morally and ethically and strive to kick those who aren't out of the system. This includes fighting against young and old scientists alike while simultaneously raising the stature of ethical scientists. Over time, the problem will be solved and we'll be able to have more discussions about scientific work rather than some jackass lab director who hears about a finished project for the first time and decides he needs to be first author.

Dear All,

I have read an article on RG. Number of authors: 1636. Number of pages: 9. IF value of the journal : 7.94. IF value per author: 0.0048.

I think the list of autors was longer than the article itself.

How much idea, work have been contributed to this article by a single person?

@Andras. Were the authors listed separately, or were they part of a collaboration or single organization. I like the way that the group at CERN publishes their work (using the name "The ATLAS Collaboration" to represented some 3000+ contributors).

Dear Patrick,

OPAL Collaboration stands in front of the names. The person I have observed has had about 600 articles and over 90% of this are similar collaborations where one needs about 5-10 minutes even to find the name of an author.

quite interesting, but this is not possible for biology, I think

My question is:

How much idea, work have been contributed to an article like that by a single person?

Many academic staff desperate to have a quite number of publication required to get promotion and for that they are willing to pay to be free rider in a published paper. I basically agreed with Ernesto that a co-author "...should be able to defend and understand all the main aspects of the paper." However, this can be done in hours to memorize and seek for content understanding from the main author without having any real contribution to the paper. Other similar question is that how much we have to contribute to be considered as co-author in academic publication? You also may find and discuss the issue here: bit.ly/1hL3YUK

Dear All,

there are some specific rules to be defined author of a scientific work.

We just wrote an Editorial about authorship problem (wth particular regard to Nuclear Medicine and young researchers). https://www.researchgate.net/publication/261436826_Authorship_problems_in_scientific_literature_and_in_nuclear_medicine_the_point_of_view_of_the_young_researcher?ev=prf_pub

Article Authorship problems in scientific literature and in nuclear ...

To me it seems impossible to encode all the relevant information regarding each contributors contribution to a scientific piece of work in to a list of authors. I propose to specify each persons contribution in an online network, .e.g, by using the Author Contribution feature in ResearchGate. I have argued for this in this commentary

https://www.researchgate.net/publication/231225109_It_is_time_for_full_disclosure_of_author_contributions

Article It is time for full disclosure of author contributions

Co-author can be any one of those whole helped in achieving research targets of formulating these, funding persons, helping in preparation of manuscript and critical analysis or updation / reviewing and editing before submission, providing samples or checking paper very critically for early acceptance, support you morally and logistically in formulating and finalizing the paper, performing statistical analysis.

Dear Prof. Linas Balciauskas

We should first think about whom to acknowledge then whom to associate as coauthor will be much clearer.

Dear Afaq

now please think about hypothetical situation. You did following services: collected material and presented it abroad; in second case you did statistics; in the third situation you was main GIS person; etc. So, it was planned you will co-author these papers. You planned your time, you did the job. After all you ger three ackowledgements. Is this OK with you?

Dear Prof. Linas Balciauskas

I am fully convinced with your view point.

Dear Lenas, a good question. The above contributors has given many good points. I think it will be wise to follow some guideline to decide who should or shouldn't be coauthor of a scientific paper. Many time the criteria may differ journal to journal also so we need to follow the standard guidelines...

Here I am attaching some guidelines which may be helpful in this discussion...

http://louisville.edu/environmentalgenomics/bioinformatics-biostatistics-computational-biology-facility-core/guidelines-for-determining-co-authorship.html

http://provost.yale.edu/academic-integrity/authorship

http://wustl.edu/policies/authorship.html

In my opinion co-authors are the ones who have contributed in the research work in the form of research review, data collection and analysis, delivery of interventions (if any), search of research literature or editing menuscript.

Dear Prof. Linas Balciauskas

Coauthor is naturally not the creator of the theme of the paper or designer of manuscript. However, coauthor directly contributes in delivery of the the theme or design of the paper.

The International Committee of Medical Journal Editors (ICMJE) recommends that all four of the following conditions must be met for authorship:

* substantial contributions to conception and design, or acquisition of data, or analysis and interpretation of data

* drafting the article or revising it critically for important intellectual content

* final approval of the version to be published

* agreement to be accountable for the accuracy and integrity of the work

Among things that do not justify authorship are:

* acquisition of funding

* collection of data without intellectual contribution to the paper

* general supervision of the research group

http://www.icmje.org/recommendations/browse/roles-and-responsibilities/defining-the-role-of-authors-and-contributors.html

Potential contributors can be co-authors, even a person / researcher with fine tuning, editing and improving the manuscript can be an author for the best

P Lease refer this link:

http://www.icmje.org/recommendations/browse/roles-and-responsibilities/defining-the-role-of-authors-and-contributors.html

(Defining the Role of Authors and Contributors)

Thanks for the resource, Selvaraju Sivamani. Previously, in this discussion, we debated a similar (or maybe exact) list. In my opinion, this list remains very vague in scope. As I asked before, what constitutes a "substantial contribution"? How is this judged, and by whom?

Also, Linus...since ResearchGate does not allow original posters to close a discussion topic, might it be possible for you to put the phrase [TOPIC CLOSED] in the title? This is an old thread.