Is it appropriate to assign two PhD students to do the same research topic?

Question

I have asked one student A to do a research topic A. He used 2 years to learn a skill. I ask him to apply this skill to a new data and research topic B recently, however the progress is still bad even if 2 years experience on learning this skill.

Is it appropriate that I ask another student B to learn the skill to look into the research topic B?

My feeling is that student A do not really try hard, e.g. absence without reason. So I suspect that the skill is not that difficult but instead student A is under-performing.

P.S. the reason why I have such feeling that the skill is not that difficult is that long time ago I ask student A to try something, and after 3 weeks he was still not able to debug. Then I asked another RA to have a try and she basically finished the framework in 3 weeks.

Answer 1

It is appropriate to ask a student to learn a skill. It is not, however, appropriate to put doctoral students in competition with one another. I'd term it unethical in an advisor, actually. Make sure that you aren't getting in the way of success of the first student.

You should, of course, speak in person with person A to see what the blockage is and if their "absence without reason" has an appropriate explanation.

Answer 2

Don't engineer a situation where your PhD student is likely to be "gazumped"

Even the best PhD students are quite slow researchers compared to experienced academics, and as such, they always face the hazard of being "gazumped" by another researcher taking on the same topic and doing it faster and better than they can. This is an omnipresent hazard in research but it can be mitigated by choice of an appropriate topic and avoidance of obvious clashes. For a struggling PhD student, the last thing they need is for their supervisor to be engineering a situation where there is a high likelihood of this occurring from a competitor who is a superior researcher to them.^

One of the responsibilities of a supervisor towards a PhD student is to help them identify a viable research topic where they can carve out an area for themselves that is likely to yield publishable work. This typically involves helping the student to find a topic where they can work for a long period of time (because they are novices who are slow) with low likelihood of being "gazumped" by superior researchers. Of course, we can never guarantee that there isn't some academic out there that we don't know about researching the same topic, but we can at least avoid engineering a situation where this adverse outcome is likely to arise.

If your struggling student is at a point where they are incapable of completing the work themselves, and it is at a dead end, then a possible second-best option would be for them to take on a co-author to help. In such a case, it is as least possible that there could be value in a collaboration between students, so long as they both contribute and both get something out of it that progresses their candidature (e.g., a joint publication). Even in this case, you would need to be very careful to manage the collaboration to ensure that both students are learning appropriately, making genuine contributions to the work, and both are progressing their candidature with the work (rather than one undermining the candidature of the other).

---

^ It might help to think about this another way. Every PhD student I've ever supervised has been a worse researcher than me at that time (which is why I'm teaching them and not the other way around). In each case, I could easily take their topic and do it myself much faster and better than they can, and publish the results on my own before they have made enough progress to do anything. Would that be fair? Would that be reasonable supervisory practice, or might it be considered to undermine their candidature?

Answer 3

In addition to what the respondents have written, consider the long-term impact on your research group. Even if it's not your intention to foster competition, your plan amounts to pitting your PhD students against each-other for first authorship. Aside from being unethical, that disincentivizes your more experienced students helping out new or struggling ones. It also erodes trust with your current students- regardless of what student A is actually doing, the message is that if you fall behind, watch your back. This does not sound like the path to a better-functioning research team.

Answer 4

You are forgetting one crucial thing: they are your PhD students and, therefore, they have to produce a thesis by the end of their 3–5 years of work. The most important thing you need to worry about is if giving the two students the same project will result in one student not being able to include the work they did as part of their thesis.

For instance, I was friends with the students in a research group who all worked simultaneously using the same machine, running all the experiments together. They were, however, very clear on whose project each experiment was for, and who was to be the lead for each series of experiments. All of them were authors on the papers, but only one of them would be the primary researcher, if that makes sense. Consequently, even though everyone contributed to each experiment, only one person could include the particular series of experiments in their thesis. This was only possible because their experiments only lasted a few hours each day, and they could record all experiments for each individual system (project) in less than 4 weeks. That does not seem to be the case here.

It sounds like you are suggesting that both students work on the same project where no one student is the lead. What this might mean is that Person A, who has spent 2+ years working towards Project 1, may face the threat of having Person B swoop in and take this project away from them because, if Person B were to complete Project 1, it would only be right for them to include it in their thesis (meaning Person A couldn't).

You could, on the other hand, put Person B onto a Project 2, which is closely related to Project 1 and could, eventually, feed into it, but it would have to be such that they were sufficiently distinct enough for each to be able to be included in separate theses. That way, Persons A and B would eventually be able to help each other out and have someone to bounce ideas off (which might increase the productivity of Person A). It would only be ethical to put another person on Project 1 once Person A has already finished up their thesis.

If you're worried about Person A not getting anywhere in their project, I would suggest that you talk through the problem with them and point them in the direction of the answer. If they're not stuck, and they're just lazy, give them deadlines (better yet, have them come up with their own fortnightly goals) and let them know they need to do more. If this has not already come up during their performance reviews, then I'd question what it is you think being a supervisor actually entails.

Answer 5

The other answers deal with the "overlapping topic" problem as you asked. I give you another answer which you may not like to hear, and which you did not ask for:

If you arrive in this situation something has gone wrong in the communication between you an student A. If you have the feeling that the progress is so slow that it does not make sense to pursue it in that context, you need to communicate that to student A. Do not try to avoid this unpleasant conversation, which may go up to the point of you stopping the project with the student by pitting the students implicitly or explicitly against each other, I have seen labs poisoned by something like this.

It could be that you find a split topic-wise, and that student A actually profits from a collaboration where they do a smaller part, and the whole project is done together, but communicate this openly.

Answer 6

PhD students are not just labor for their supervisor. Actually, as their supervisor, part of your duty is to ensure that they’re happy with their subject and working conditions, and to help guide them towards topics that can appeal them. It will be difficult to get successful students otherwise.

Therefore, in the light of your question, you should ask student A if they’re willing to pursue project B, or if they would prefer to focus on another topic and have someone else working on project B (it is not clear from your post whether project A is finished or not, maybe they’d just rather continue to investigate this topic ?).

Do not place PhD students in a situation of competition, nobody will gain from that. Place them in a situation of cooperation. If student A is not interested in project B, they may be happy to have another student as a colleague, and they will be very helpful in onboarding the new student anyway — especially if their work serves as background for project B.

Answer 7

I agree with the other answers that this is generally a very bad idea. One possible solution would be to hire another PhD/postdoc with a similar but nonoverlapping topic. Having this extra expertise in your group might help your "slow" PhD progress faster. One possible downside: by the time this other person will be up to speed, student A will be quite far along their track and there may not be enough overlap.

Answer 8

One can approach any research topic from numerous angles. Two students working on the same topic will almost always produce different results IF AND ONLY IF they are working independently.

The main question here is if this is a time sensitive issue. Meaning, your lab or group must produce something by a specific date because your research grant sponsor is asking for results to justify further funding.