Co-authorships in academia

by Eszter Hargittai on October 5, 2004

Few academic institutions put anything concrete in writing when it comes to promotion and tenure review so it seems an informal discussion on a blog about the topic will be as informative as most other opportunities to consider the issues.

I have been pondering the pros and cons of co-authoring articles during one’s junior faculty years. How does a co-authorship count toward promotion and tenure? Obviously the answer is going to depend on a myriad of factors, but a discussion may still be interesting and illuminating. I realize that in some fields co-authorship is more the norm than the exception. In most lab sciences one rarely sees a sole-authored publication. But in the social sciences – the home discipline of several CT authors – it is less common. Since there are tenured faculty around here who have likely participated in promotion and tenure reviews, I would be curious to hear about their experiences. Of course, others are just as welcomed to contribute their thoughts.:)

What types of publications are possible regarding authorship?

Sole authorship. These publications have one author. These papers are the most straight-forward in terms of whose work they reflect and are highly desirable, it seems.

Co-authorship with senior faculty. These papers are co-authored with senior scholars. Junior faculty are often advised not to have a CV solely consisting of papers written in collaboration with senior scholars. It is important for a junior scholar to show clearly his or her distinct contributions to a field and by co-authoring with senior scholars, some will be inclined to dismiss the work as that of the senior researcher. This is not necessarily the case, but it may happen and so a junior scholar will likely want to diversify.

Co-authorship with peers (other junior scholars). These pieces are co-authored with peers of a similar tenure rank. If both authors are known for the particular topic of the paper (or it is an area new to both of them) then the contributions will likely be considered equally. Neither author will be given more credit than the other for the publication so assuming equal contributions to the project, credit will be assigned reflective of actual work contributed.

Co-authorship with students. On these publications, the junior scholar becomes the senior author due to his or her seniority as compared to the student co-author(s). None of the authors may be well known so credit is likely assigned equally. It is unlikely that the junior scholar will be given less than half the credit for the work (or more generally speaking, in the case of n co-authors, less than 1/n credit of the work).

I was “raised” in the tradition that if students participate in a research project actively and contribute to the writing of the paper then they receive co-authorship. I was also “raised” in the tradition that if a graduate student did most of the work on his or her own paper then the faculty member advising the paper does not put his or her name on the publication as a co-author. It is the graduate student’s work, the faculty member played an advisory role. This is clearly very different not only across disciplines but even across departments or simply individual scholars. I mention this to clarify that in the preceding paragraph I am not referring to projects of graduate students.

So how should a junior scholar proceed? Pursuing only sole-authored publications is not always viable because some projects greatly benefit from – or are only possible with – multiple contributors (e.g. projects where the different members of the project bring very different skills to the table). It is also easier to pursue different interests if collaborating with others partly because there is some division of labor, and depending on the co-author, there is regular prodding to advance the project.

If given the choice of writing a sole-authored piece (one that the junior scholar could pull off on their own given enough time) or bringing a graduate student on as a co-author (making the project more likely to reach a publishable state thanks to the division of labor), which is the better approach? Is it better to wait to finish a piece (given another dozen pieces in the works with varying levels of priority) and have it be sole-authored, or is it better (or just as well) to bring on a collaborator who will help get the piece out quicker?

That’s just one possible question. There are lots of others, feel free to raise your own. I am involved in projects of all types described above so I have some experience with all variations. They all seem to have their pros and cons.



MC 10.05.04 at 10:18 pm

As a very junior grad student in a discipline that straddles between the clinical and social sciences, I don’t know the dynamics at all, but I’m interested in finding out the answers to these questions too. Due to the nature of the work I did before grad school, my (very short) cv consists of mostly sole authored works. I was given the advice to make sure that I add some collaborations or otherwise it would give the impression that I couldn’t work with other people.


John Quiggin 10.05.04 at 10:22 pm

In economics, the two-author paper is probably most common. Except for the supervisor-grad student case, co-authors are mostly peers.

I’ve seen estimates (and it’s consistent with my experience) that each author gets about 70 per cent credit for a joint paper, which makes co-authorship an optimal strategy in most cases.


Fabio 10.05.04 at 10:33 pm

In theoretical particle physics, pooling seemed to be the preferred strategery for maximizing job prospects. As long as the number of authors was small enough to fit on the cover page, any particular author could claim close to 100% credit when giving a talk on the subject.


martin 10.05.04 at 10:34 pm

In a nutshell: no matter what you do, you can’t win! But you can obtain tenure and get promoted

My experience of P&T committees is that you cannot write any hard and fast rules. Each year, my department elects a new committee. How that committee views the issue of authorship reflects the prejudices of the individuals involved.

I’m in biochemistry. In times past, papers from one’s own lab were what mattered. Nowadays, the granting councils favour collaborative research. I’m not convinced that individual faculty members have really thought about the remaifications for judging P&T files.

It is important that one submit a body of work that has clearly been driven by the applicant. It would be rare to see a senior scholar/junior scholar situation in the sense of the senior scholar mentoring the junior. That would not be viewed well. As long as there is a balance of work – some papers clearly the major effort of the applicant and some papers the work of broader collaborations, one is usually OK.

For the most part, papers are written by graduate students who get first authorship. The faculty member, who probably wrote the research grant to get the money, will be the author of record – usually cited last.

You can probably find details of P&T matters at most Canadian universities by searching for faculty union websites at the respective universities.


eponymous 10.05.04 at 10:43 pm

Maybe helpful, maybe not…

While in graduate school, I was fortunate enough to be able to sit in and observe faculty meetings where promotion/tenure issues were raised. Specific instances with respect to individual cases I was not allowed to observe; however, during those times I was able to sit in and listen/observe, I was able to discern the following:

Authorship in and of itself is only important with respect to the journals you are published in. It could very well be the case that all of your papers are sole authorship. But if too few of them are not published in the “right” journals, then gaining tenure or promotion may be difficult.

Bottom line – if you belong to a department that prides itself in having faculty publish in top-tier journals, then I wouldn’t be too concerned with the sole/multiple authorship dilemma. Just make sure that the bulk of your articles (whether sole or multiple authorship) are published in the top-tier journals.

NOTE: The above pertains to one of the top geography departments in the country – make of it what you want.

Maybe the following schematic will help better illustrate my observations:

1) Publish a lot and bring lots of money into the department (via research grants);

2) If you can’t bring in lots of research money, then publish a lot;

3) If you can’t publish a lot, then make sure that you publish in the top-tier journals and that the bulk of your published work in the top-tier journals is sole authored;

4) If you can’t publish all of your work as sole author in the top-tier journals, then publish with others in the top-tier journals;

5) If you can’t publish in the top-tier journals, then make sure you publish as sole author in second-tier journals;

6) If you can’t publish in the second-tier journals as a sole author, then publish with others in those journals;

7) If you can’t publish in second-tier journals, then publish a book (or text) as sole author;

8) If you can’t publish a book (or text) as sole author, then publish the book (or text) with others;

9) If you can’t publish, then make sure you can teach;

10) If you can’t teach, then start looking for another job :)


Barry Freed 10.05.04 at 10:43 pm

I’m in the humanities, Islamic studies, and still a few years from going on the job market. But here are some thoughts:

I really don’t think that it would be a good idea for junior faculty to co-author a paper with a graduate student. When the time comes for tenure review it could lead to questions being raised about who was responsible for the preponderance of work. Though if you’ve got a slew of publications I don’t imagine one such would hurt you. But if your pubs are few, and there a number of such articles. That can’t bode well.

[I do think you should check actual published articles though]

I tried to think of any co-authored articles in my field and I was about to say there aren’t any but one came to mind. The thing about is that both scholars were well-established and given the topic and their respective expertises (?)it made sense for them to collaborate on the article. [for the curious, from memory the article is “Neo-Sufism Reconsidered” by R.S. O’Fahey and Bernd Radtke and is not without relevance to contemporary concerns vis a vis political Islam].

There may be other’s but the phenomenon is fairly rare.

We have another issue affecting us in a similar way: Translation.

For tenure review purposes translations are given very little weight. But in Islamic Studies, particularly in some of the more arcane and technical areas like theoritical Sufism good translation can almost constitute a study in it’s own right. Also given the vast quantity of material that has still to be looked at, if you choose a source text randomly you’d be almost guarnteed to be breaking new ground. [I’ve seen academics in English and Classics start salivating when they find out just how abundant our MSS sources are and how little has been looked at, and much even yet to be catalogued properly.]

My case in point here, though it didn’t adversely affect the proffesor’s tenure, is the work of William C. Chittick. Especially “The Sufi Path of Knowledge: Ibn al`Arabi’s Metaphysics of Imagination” and the follow-up “The Self-Disclosure of God: Principles of Ibn al ‘Arabi’s Cosmology” these are texts that only a few handfuls of people in the world are capable of reading with any understanding [and that includes native Arabic speakers].

It’s truly unfortunate that such work is regarded as translation rather than a study. Of course the effect this has on knowledge of Islamic thought in the West can only be deleterious.


Barry Freed 10.05.04 at 10:59 pm

eponymous brings to mind another very important point.

This is based on very close observation (mostly a spouse, since separated):

It matters a great deal what kind of institution you are dealing with. You’ve got to know the values of the school and the department and play to them. Some institutions, the top research Universities and those aspiring and/or posing to such status will naturally place a very great emphasis on having many publications, the more prestigious the journal or publisher the better [In my field this has the unfortunate effect of far too many useful books that would find a wider audience being published by Brill and costing $100 and up.]

Many schools value teaching more, and some place a great deal of importance on fostering a sense of community. Here the importance of publications seems to me to be just equivalent to another item to be marked off on the checklist. If you’ve got the required number then great, no fuss. More important is how well do you get on with fellow faculty members? Are you a good teacher? Are you well liked and regarded by your students? Do you add something to the community, participate in some way, [preferably non-blood-letting]?


J. Ellenberg 10.05.04 at 11:12 pm

In math, John’s “70 percent rule” also applies — I’ve heard it as 75 percent of the credit to each author of a two-author paper. Also, in math the authors are always listed alphabetically and it’s considered very bad form to assert any kind of heirarchy among the authors.

In my experience, the intellectual benefits of having a good collaborator far outweigh any “credit penalty” you might undergo for co-authoring.


Barry Freed 10.05.04 at 11:48 pm

j. ellenberg:

In my experience, the intellectual benefits of having a good collaborator far outweigh any “credit penalty” you might undergo for co-authoring.

Thanks for that. Now I know what all the fuss is about having a high Erdos number. It’s far more than a geeky mathematicians in-joke but is actually reflective of how whuffie functions in the field.


Cryptic Ned 10.05.04 at 11:49 pm

You’re not going to see a self-authored paper in the lab sciences, unless it’s purely theoretical (e.g. Hawking) or it’s a review article by someone who’s one of the top 5 experts in a field (e.g. “Recent Developments in Cytotoxic T Lymphocyte Trafficking”).

It takes more than one person just to a) do research and b) get the supplies. If you want to hear what one person and one person alone has done recently, you have to see their posters and presentations at conferences, not their published work.


djw 10.06.04 at 12:22 am

I’ve been told by a friend in a good-but-not-great R1 that a co-authored paper or two (with in-house graduate students) would be a big asset to his tenure applications, because the department is worried about the marketability of its PhD students, in part because none of the faculty ever publish with them. I’ve also seen a few job ads in my field (political science) at top tier liberal arts schools that say something about incorporating undergrads in research and writing as a desirable characteristic. I interpret that as schools searching for a leg-up for their students when they apply to grad school. I served on a graduate admissions committee a few years ago, and it seems to me that strategy works pretty well.


John Lott 10.06.04 at 12:24 am

My strategy was to fabricate data and then co-author with imaginary graduate students. It worked great. For a while.


John Quiggin 10.06.04 at 12:44 am

Plagiarist! You stole that idea from sir Cyril Burt!


Mary Rosh 10.06.04 at 12:52 am

Plagiarist! You stole that idea from sir Cyril Burt!

I won’t stand for that kind of criticism — John Lott was one of the best professors I ever had in Graduate School at Sockpuppet University.


harry 10.06.04 at 12:53 am

I sole authored all pre-tenure articles, and that is the standard thing in Philosophy, I think. I also think that in Philosophy each author gets less than 50% credit for a co-authored piece (in people’s minds, at least). This is fine if you are writing something that you simply could not have done, or could not have done as well, alone (as is the case with my own pieces in progress co-authored with another political philosopher) even given the time, and also if you are learning from them (as is also the case), or simply having a good time (also true).

However, if you co-author with someone in another discipline you get all the credit (100%) because philosophers simply can’t imagine that a non-philosopher could have contributed anything that was of value. That may be an exaggeration, but you get the idea.

Philosophy likes to think of itself as closer to the harder social sciences and the sciences than to the rest of the humanities (journal articles over books, rigour over mush, you know). But the overwhelming tendency to sole authorship is one respect in which it resembles the other humanities more.

One last thing. Aim for what are regarded as top-tier journals. Maybe I am particularly self-effacing but I wrote a paper targetted for a very obscure journal pre-tenure, and only sent it to the top-tier journal which took it after a senior colleague laughed at me for what he wrongly took to be my modesty, and all but submitted it on my behalf.


ProfGoose 10.06.04 at 1:37 am

There are no doubt departmental and university norms, if not codified norms, at any institution. My experience (as a social scientist with a “hard” science bent) is that the situations are completely idiosyncratic at different institutions/departments.

For instance, I have just moved from an institution where co-authorship was frowned upon, but not held begrudgingly against someone in tenure cases, depending on volume.

In contrast, here at my new institution, co-authorship is lauded and incentivised. As someone who appreciates the social aspects of co-authorship (bouncing ideas off of others who care about the project primarily. I agree with J. Ellenberg…), I am ecstatic about my new situation and very excited that the department and university are accepting of my way of being.

Of course, it doesn’t hurt that I publish my co-authored pieces in the top two tiers of journals in my discipline…


schwa 10.06.04 at 6:36 am

I have nothing of intelligence to say on tenure generally, since I’m still a lowly grad student, but it’d be very interesting to see a disciplinary census of co-authoring practices.

In my discipline (comparative politics) the sole-authored paper seems to be the norm, but sometimes — and rising well above “sometimes” in papers which lay the quant. on thick — you get two authors. Higher numbers than that seem fairly rare.

On the other hand, international relations seems to deal much more in multi-author papers.


andrew cooke 10.06.04 at 12:14 pm

i’m surprised at the amount of variation here.

in astronomy, most papers have more than one author and the first receives most credit. it is important to publish in the major journals (there is a known hierarchy, but also an understanding that certain nationalities may favor certain more local journals because they have lower costs).


Doug Turnbull 10.06.04 at 2:21 pm

My background is in experimental physics/electrical engineering. In that field, almost all papers are co-authored, just because of the way things work. Usually, the way it’s done is the person who did most of the actual work (grad student or post-doc) is the first author, with subsequent authors ranked according to their contribution. The exception is the professor/lab head who is put on last.

The first author and the professor get full credit for the paper, with everyone else getting partial credit.

As others have noted, for tenure decisions the stature of the journal is critical. One article in a top tier journal is worth 3 in obscure ones.

For a research university, I’d pretty much agree with eponymous’ rankings above. A pre-requisite is the ability to do solid, respected work. Given that, the most important factor is the ability to bring in money. Then the number of publications.

The exception is the rare soul who actually makes a major contribution to a field. Then, the caliber of the work outweighs the number of publications. (ALthough if you do work of such high caliber, it’s usually not hard to get money and lots of publications.)


harry 10.06.04 at 2:48 pm

My godfather is a Nobel prize winner in Medicine (he’s a biochemist). Not to name drop or anything. But I was really struck, when looking at his Nobel file (proudly shown to me by his wife, my godmother) that there were just one, maybe two, sole-authored publications, out of the what seemed like hundreds. I don’t know anyone in the humanities who could do that.


Ross 10.06.04 at 4:10 pm

An outstanding discussion.

I just completed the T&P process last fall (successfully, thank goodness) at my medical school, but my circumstances — as a law/policy scholar in a clinical/bench science world — made the water even muddier, both for me as I came up through the system, and for the T&P committee, who was forced to rely heavily upon my chairman’s letter and my outside letters for guidance on how “well” I did in the publication area.

I had a mix of a couple single-authored pieces in law review journals, and co-authored pieces and one treble authored piece in law review journals and peer reviewed policy, ethical and scientific journals, and a co-authored editorial in Science, which they call a “magazine.”

As partial guidance, our T&P committee could look to our law school, which requires two or three single-authored major law review publications, with at least one in a school’s primary law review journal. Most law schools frown upon co-authored work in the pre-tenure period, and importance of journal rank varies from institution to institution.

In the medical field, most everything is co-authored, so it is most important for a young scholar to be high in the list of authors (1st or second, ideally), and, as someone mentioned earlier, the final spot is reserved for senior scholars, the same way “and Jane Seymour” might appear as the last credit for a TV movie. Placement can make a difference at some institutions, but volume goes a long way too. I know of one department chair who said he would not write a letter on behalf of someone going up for tenure if they didn’t have at least 15 publications under their belt. Research dollars are important to the basic scientists, so while my field is not big on research dollars, I was fortunate enough to find a research collaborator with whom to partner as co-Primary Investigators on a grant.

Not sure how helpful (if at all) this is for your specific case, Eszter, but there it is.


eudoxis 10.06.04 at 5:40 pm

Whether or not to co-author strikes me as fine-tuned and luxurious strategy contemplation. Projects that can be developed for single author publication are clearly valuable but not to the exclusion of collaborations. A significant body of work is not a matter of numbers. Of course there’s no recipe for tenure, but not having babies and a lateral move around tenure time are advised.


Giles 10.06.04 at 7:34 pm

Surely the problem though is that if co authors get 75% credit, then if on average say 1/2 the papers are co authored, in academia there’s 25% more precieved credit than actual output.

In other words, in aggregate people on Selection Committees are acting as though there’s more research going on than there actually is. Might this cause an over bias towards published research in academia as a whole?


Paranoia? 10.06.04 at 9:09 pm

If I weren’t dealing with an itsy-bitsy comment box, I would type out a logistic regression equation with estimated coefficients for each of the possible authorship combinations, interaction terms allowing the coefficients to differ by discipline and, of course, a large stochastic term. But I won’t.

I will, however, offer an anecdote. I am first author on a paper that has recently been accepted (yay!) at the top journal in my field. The second (and only other) author is a senior colleague. It so happens that my senior colleague’s last name would appear before mine in an alphebetized list, so it’s hard to interpret the order of authorship as an accident of alphabetization. I have written all correspondence with the journal editor about the article. The acknowledgments on the front cover of the paper include the phrase, “direct all correspondence to [me]”. The editor of the journal (a former colleague of mine) knows exactly where I work, and he also knows that my coauthor is in the process of moving across the country.

And yet the acceptance letter was addressed solely to my senior colleague. Clerical snafu? Maybe…


John Quiggin 10.06.04 at 9:09 pm

Giles, there is indeed a bias.

One point I forgot to mention is that there’s a study somewhere showing that women in economics are less likely to co-author than men. Because of the bias already noted, this harms their chances of promotion.

There was some speculation about the causes which presumably relate to the social dynamics of a lopsidedly male discipline, but I can’t recall more than that.


Richard Zach 10.06.04 at 9:38 pm

One of the referees on a recent project proposal noted (positively) that I had already collaborated with <name of noted person in my field redacted>. It was an application to NSERC (the Canadian equivalent of the NSF), and in the field (logic), co-authorship is common. But even in other fields where co-authorship is not common (say, philosophy), writing a paper with a noted senior colleague will probably more help than hinder. After all, if you’re not any good, why would <noted senior person> write a paper with you? This may be less true, however, if <noted senior person> is your dissertation supervisor.


Claire 10.06.04 at 10:43 pm

Coupla comments.

My impression is that having some co-authored papers helps for going on the job market, it shows you can work with others (a Good Thing in a colleague) but too many coauthored papers hurt tenure chances (I’m in linguistics, and not tenured, only just off the job market, so this is mostly from observation).

Secondly, there’s another category of coauthorship in linguistics, and that’s publications with one’s field consultant(s). This isn’t all that common as far as I can see, for a couple of reasons:

. the consultant doesn’t need tenure.
. in university systems where brownie points are awarded for publications (which affect funding, in turn) coauthored publications count for less.
. there’s a perception that the consultant does less work than the linguist.

All my language materials (learners guides, dictionary supplements, text collections) are coauthored with my consultants and the analysis articles/phd etc are single authored. I would feel really bad about being sole author of a text collection where someone else told the story in the first place, someone else helped me transcribe it and a third person helped with translation. That’s a pretty large part of the intellectual component of the work! and having their names on the front is a source of tremendous pride to the consultants who worked on the text collection. Besides, coauthoring language materials with consultants can’t hurt tenure more than doing text collections, etc, in the first place …


Alex 10.07.04 at 2:06 am

Much ado about more hot air.
Huge waste of time, moneys on all of you, and paper.
Univ model is archaic and outdated.
Ruined the country.

Besides, there’s nothing much new under the sun other than some hard science research results.

Made at least three attempts to comment thusly, but got deep sixed.

Comments on this entry are closed.