When to submit to journals in grad school?

In our May "how can we help you?" thread, a grad student asks:

I'm sure you get questions like this all the time so feel free to direct me to an answer if one already exists. But does anyone have a good metric for when a paper is ready to submit for publication as an early to mid stages grad student - when someone tells you you might get a publication out of it? It's accepted at any conference? It's accepted at an APA? You just feel good enough about it!? I assume none of these things are necessary, but do people take any to be sufficient?

These are excellent questions. I feel like we have discussed these issues sometime before, but if we have I can't quite track them down. My sense is that the best thing to do here is probably to have a few faculty in one's grad program read the paper to see whether they think it is ready. What do you all think?

Comments

[Newly Dr.]

I would say that the best guide is: when your supervisor, or some other faculty member you trust, says that it is ready to be submitted.

That is, assuming you have a supervisor who you trust. I hope that most people have one. If you have a supervisor who you think does not have your best interests at heart and might suggest not submitting even when a paper is ready, then you probably need to work that out first. Similarly if you don't have a good enough relationship with your supervisor to ask them.

Of course, you might have a supervisor who is a little unforgiving and be worried that they will never be happy with your work. Then I can only suggest you get it to the stage that they think it is almost ready, and then submit it. I did this with the two papers I got published during my PhD—I knew my supervisor would probably never say that a paper was ready (or at least, not unless I suddenly got a lot better at writing them) and so I just sent them off when my supervisor was happy enough to only have fairly minor suggestions. Both papers were published, after revisions, at the first journal I sent them to (both in the top 20). Probably my supervisor was right if I wanted to get them in the top 5 journals, but I am happy with them.

To answer your specific suggestions I would say:
1. "you might get a publication out of this" is not sufficient, since this generally means that the paper might one day become a publication.
2. Whereas "you should publish this" or "this is ready for submission" are exactly what you want (from reliable faculty, not other grad students).
3. Accepted at any conference: definitely not enough.
4. Accepted at a competitive conference: still probably not, because conferences generally have lower standards. Although if the whole paper was reviewed and the conference is particularly prestigious this might be enough.
5. You just feel good enough: see comments below.

Finally, I don't think that submitting too early can really hurt you. The worst they can do is reject the paper, and they will probably do that anyway (unless your work is extremely easy to identify). So you should be bold and submit once you think something is ready. This sort of strategy is a problem from the perspective of the whole profession, because it overloads the peer review system. But until more senior members, and administrators, move to reduce the publication pressure I think it is only fair that you do what everyone else is doing and try your best to get your work published even at the risk of compounding the current problems with peer review.

[Bill Vanderburgh]

If it is a paper written for a class, send a revised version to the professor and say, "I'm thinking of working this up for publication. What do you think it needs to be ready?" If you are at the dissertation stage, and especially if it is a paper from your dissertation, work closely with your supervisor. You don't want to send out something that isn't fully ready, since that will affect how people view the quality of your dissertation (though that is in tension with the need to have some publications before you defend).

I think the advice from grad programs used to be to just send stuff out, even if it isn't ready, and then use the referee system to polish the work over several iterations. This shifts the burden of getting the paper ready from local faculty to the discipline as a whole, and possibly avoids the hurt feelings of mentor telling you something isn't good enough. This method has some "professionalization" advantages (learning to work with editors, responding to referees, etc.), too, but with the current referee crisis I'm hesitant to recommend this practice. It is better, I think, to use your personal networks to make the paper as good as it can be before sending it out for the first time.

Generally it is much easier to get a paper accepted at a conference than in a journal, so conference acceptance is not a good guide to journal readiness. But if you have a paper accepted for a conference and you have a commentator, a version you revise in light of the comments and audience feedback is probably close to ready. Send it out and see what you get back from the first journal.

The one exception, I think, is APA. Those papers are refereed to a high standard and the acceptance rate is low, so a paper accepted for the APA is probably pretty close to publishable, especially once the commentator's feedback has been incorporated.

Grad schools, faculty development programs, and research offices may have formal or informal programs for helping grad students write up their research. These are definitely worth exploring.

[Junior TT]

I would send out the paper if you think it is pretty good, as you will improve it from referee comments and also because it takes so long to get things accepted. This is what I did in grad school and my papers were eventually accepted at good journals—though I don't remember ever being advised to submit them (or not to). You should edit it first using comments from peers and mentors but I would not hesitate to send it out if you judge it's in decent shape.

Bill Vanderburgh writes, "I think the advice from grad programs used to be to just send stuff out, even if it isn't ready, and then use the referee system to polish the work over several iterations...but with the current referee crisis I'm hesitant to recommend this practice."

I'm not sure the reasons in mind here. If the reason not to recommend just sending things out is that if grad students do so it will only exacerbate the crisis, let me say that I was hesitant to write the first paragraph I wrote for the same reason. But upon reflection, I came to think that this is a disciplinary problem and not one that grad students should have to have a hand in fixing. Especially if their peers (or successful predecessors!) have already benefitted from it. It's an obnoxious practice but the stakes (getting a job) are too high to worry about it in grad school.

[R]

"I would say that the best guide is: when your supervisor, or some other faculty member you trust, says that it is ready to be submitted."

I agree with this, but don't forget to explicitly ask them! A supervisor who is otherwise good and trustworthy might still not volunteer an explicit comment "you should submit this for publication" without being asked. They might simply not be thinking about the need to submit papers for publication as a grad student (if they're a bit older and not as aware of what one needs on the job market these days), or they might assume that you're already sending it out for publication once you've received any vaguely positive comments from them (or if they're a bit absent-minded they might even think they've already told you to send it out).

[Oliver]

Bill Vanderburgh says: ‘ You don't want to send out something that isn't fully ready, since that will affect how people view the quality of your dissertation’

I can’t see how this works, as it is extremely unlikely the referee of your (rejected) paper is also likely to be your PhD examiner. The worst that can happen is your paper is rejected and therefore no one ever sees it.

[Kevin]

I think you should go for it—send out a paper, see what the comments are like. I sent out two papers without telling anyone in grad school. The first got rejected, the second got a favorable revise and resubmit. My professors would have discouraged me from sending things out so soon, but I wanted to see what it was like.

[lucky publisher]

I do not think there is a linear comparison to be made between conference acceptances and publication. I had one paper rejected from multiple conferences (even from some graduate conferences), but the very same paper got accepted at a top-5 specialist journal in my field and I am sure there are many conference acceptances which do not turn into a publication.

I think the moral is that conferences operate with many auxiliary demands over and above the quality of the work (diversity of participants and topics etc. etc.) where the same may not be true for publications.

[APA program committee member]

Just here to disagree with BV's claim here: "The one exception, I think, is APA. Those papers are refereed to a high standard and the acceptance rate is low, so a paper accepted for the APA is probably pretty close to publishable, especially once the commentator's feedback has been incorporated."

--The acceptance rate is high compared with many conferences and workshops. (Though it is low compared with conferences that accept most things, of course.)

--The refereeing is often based on quick snap judgments, each paper is only refereed by one person and they have a large pile of them (and, at least in the division I've refereed in, you simply assign a number, so there is no need to justify your decision).

--When I've refereed for the APA (won't speak for others) I'm not looking for publishable papers. This is partly because the papers are so short (I don't see how I would know if the longer/fuller version of the paper is publishable), partly because I am refereeing a much broader selection of papers than what I would agree to referee for journals (including many on topics that, while in my AOSs, I don't work on and don't know the literature on well), and mostly because I look for papers that would make good conference sessions rather than good journal articles. I think there's a big difference between the two!

In general I agree with "lucky publisher"--conference acceptance is not a good guide to publishability at all (and vice-versa), for all the reasons they state but also just that what makes a good conference paper is different than what makes a good journal article.

[2 sense]

I think you should have reason to believe that your paper is an original contribution to a debate. If not, then it will just clog the system.
What you must resist is just sending your term papers out to journals. I see some of this stuff, on my end as a referee. (at least, that is my best guess).
And, unlike what some seem to say, papers accepted at more selective conferences are good candidates to revise and submit to a journal. They have already been veted. But it is imperative to take the paper to the next level, and make it more scholarly.

[non-tt faculty]

I think there is merit to work with supervisor's approval. It is extremely unlikely that papers are accepted as they are. I found, at least through my first three papers, that my supervisor's advice on how to do R&R was extremely helpful.