Should academics allow student to submit a possibly flawed paper?

Question

I let a student associated with me (but who is not really my PhD student) submit a paper to a fairly-local meeting (some people do come from abroad). Because I'd been tasked with supporting the student, I let him put my name on the paper and also helped with the presentation, but I found the content unconvincing. However, he was exploring an area that I knew little about and had little interest in, so I'd hoped he'd get reviews & (if accepted) discussion that would give him more guidance than I have been able to.

In fact, no one seemed as sceptical as I was of the work, and it got into the meeting and was presented both in our department and at the meeting with if anything positive comments.

However, yesterday I got an email from a postdoc who was cited in the paper (the student compared himself to the postdoc's work, the meeting's papers are on a website, Google Scholar alerted the postdoc) and was incensed at the paper's low quality and inaccuracies. I think the postdoc is being a bit paranoid, but is broadly right on the technical issues. I have also previously noticed that one of my more-successful colleagues I've been collaborating with recently had a less tolerant attitude towards student publication than I do.

I'm wondering if I should put more effort on quality control. The cost would be possibly stifling a student unnecessarily if I'm wrong, and allowing fewer students to have fewer presentation experiences since I'm already working flat out keeping up with giving feedback on their journal articles & dissertations.

Answer 1

In my opinion, since you aren't the supervisor, it's not your role to stop the student from publishing. If you are an expert in the subject of the publication and you don't think it's strong, you should advise them not to publish and explain why, but that's as far as you should go.

However, you "let him put [your] name on the paper" -- a paper which, by your own admission, you don't know much about! Regardless of whether the co-author is a student, and regardless of the quality of the paper, the behavior you're describing is wrong. In my field, it's actually forbidden by at least one of our professional societies:

All the authors listed for a paper . . . must have made a significant contribution to its content.

This is generally construed as meaning you should have done a significant part of the research and a significant part of the writing.

Now, violating this rule for a paper that you don't even have a high opinion of seems not only unethical but foolish.

I realize this is a harsh-sounding answer, but you have posted the question anonymously so I hope you don't mind me being frank.

Answer 2

I think we can distinguish two situations here:

1. The work is not convincing because it is still preliminary. In this situation, even if the results might prove wrong latter, I would see no objection to send the student to present a talk or a poster about it, as long as the preliminary nature of the work is clearly stated. I might give him the opportunity to meet other researchers, find new ideas or even build collaboration based on these first results.

2. The work is not convincing because the results are flawed, the protocol is not robust or the techniques might not be adequate. In such situation, as @ff524 said, I would advise not to let the student present the work in public meetings. It will be unproductive both for the student and the advisor (as you experienced).

Answer 3

As a mathematician, I had the same first reaction to this question as several other answerers and commenters: It's wrong to be a co-author of a paper that is just someone else's work. But I need to temper that reaction with the fact that other fields have rather different standards. In particular, in some (maybe even all?) of the experimental sciences, it is standard practice for the head of the lab to be a co-author of everything that comes out of that lab, whether or not the head actually did any of the research or even understood the research. As far as I can tell, the rationale for this is that the head of the lab gets the grants that make everybody else's work possible. The question here suggests that this sort of thing might be involved here ("I'd been tasked with supporting the student"), so co-authorship might not be quite as crazy or as unethical as it looks to a mathematician. Even in the experimental sciences, though, the head of the lab is (as far as I know) expected to make sure the work is good (and is held responsible if it is not).

Answer 4

Sorry to break the "don't answer your own question" norm here, but I want to float an answer. First, I did contribute to the paper (suggested direction of research) and the system presented does work. My issue was whether this new system was a real academic contribution, which I left up to the academic process to determine.

Overnight, the postdoc who initially complained a) admitted that the idea was great if not entirely well executed and b) told me he's found a new prestigious collaboration writing a better paper. In my mind, this is how academia is supposed to work. So my proposed answer is "yes". I leave it to you folks to vote up or down.

Answer 5

My answer is, to what end? What purpose are you serving by allowing such work to be published? The community does not benefit, as the work is subpar and (apparently) flawed. The student does not benefit, as (a) the experience is different from actual publishing with more rigorous review processes and (b) their reputation is tarnished. You receive no benefit for the same reasons.

The only possible benefit I see is that the student gains experience in informal writing and presenting, which they can already gain through group lab meetings without the possible repercussions relating to their reputation. In short, this appears on all fronts to be a pretty bad practice.