I'd appreciate any resources describing how 'hands-on' an academic supervisor should be.

I've found in my own supervision that it has varied from quite high-level guidance to quite detailed guidance. For instance: at a high-level I've provided general project direction, advice on experimental design, and suggestions on how to present results; at a more detailed level I've helped install software, reviewed, edited and written code, and providing extensive reading lists.

There are some already resources available on this topic, e.g.

  • This question: addresses whether meeting every 3-4 weeks in a group meeting is sufficient (which appears to be at one end of the scale); and this question similarly deals with a hands-off approach to supervision.
  • This response: provides links to resources which provide some helpful insights, e.g. this article states that a trait of the worst supervisors is that they are never around, and at the other extreme this article highlights that "it's best not to have an overly hands-on adviser because that can handicap your future career" (based on this study).

However, I think it would be helpful to share further resources / guidance specifically on how hands-on a supervisor should be, e.g. ranging from one 1-hour meeting per week/fortnight, to helping with the mechanics of research such as identifying literature and writing code.

If you are aware of any such resources in addition to those outlined above then please do share. Thanks very much.

  • 11
    You seem to assume that there exists some ideal amount of hand-on a supervisor should be. There isn't. It varies wildly and while too extreme cases can be bad in general there is a big range of levels than can work perfectly fine for suitable individuals.
    – quarague
    Jan 2 at 12:55
  • Note that "how hands on to be?" and "how often to meet?" are not the same question. In other words, hands-on-ness is not a 1-dimensional scale. For instance, you could have very frequent meetings, during which the student reports what they have done recently, what they are planning to do in the near future, and what difficulties they are encountering, without being "very hands on"; or you could have somewhat less frequent meetings, but be super directive and tell the student what to do, and that would be "very hands on".
    – Stef
    Jan 3 at 18:26
  • I want to point out that the language in your question is already biased - I am not the supervisor for my PhD students - I am their advisor. Jan 3 at 20:18
  • Thanks for the very helpful comments. I agree that there is likely not a 'one-size-fits-all' answer, and that the frequency of meetings is not the sole determinant of how 'hands-on' a supervisor is. I also found the point about terminology (supervisor vs. advisor) very helpful. At my institution the terminology is provided for us (supervisors and advisors are distinct roles). Jan 4 at 6:12
  • 2
    I'm voting to reopen the question. Questions about the efficiency of supervision approaches are factual and could, in principle, be answered with data and citations (even if the current answers don't have that). Jan 4 at 14:45

3 Answers 3


In an efficient supervision relationship, the degree of "hands-on" supervision is tailored to the concrete needs of the student.

A typical experience is that the PhD student at the beginning of their PhD requires relatively much hands-on supervision, since they are inexperienced at what they're doing. Over the course of the PhD, they acquire the necessary skills and mindset to develop their research in a more independent way and then require less hands-on supervision.

Generally, one supervision meeting per week seems a reasonable default that can be flexibly adjusted to the specifics of the situation.

There might be some field-specific differences involved. In STEM disciplines, it can take some time for students to become familiar with certain equipment and techniques, whereas in the humanities, it seems to be more common to expect a certain level of independence from day 1.

  • 3
    There is also a difference between supervising the individual students and shepherding the group. It takes both to make a high performing team. The group meetings show the newer folks what the bar is, and the individual attention mentors them to get there.
    – Jon Custer
    Jan 2 at 13:30
  • 2
    When I first became independent I tried to model my prior groups, which were world-leading mega groups. It did not work. Because frankly those groups were 50+ people, with maybe 25+ being very high achieving members, which is more than enough to cover for those which were not. I realized that a more tailored approach to individual student needs and helping them overcome their specific barriers led to a much more productive medium (10-20 member) group. It also led to more instances of weaker students becoming much stronger rather than falling through the cracks which was more common in my past.
    – R1NaNo
    Jan 2 at 19:07
  • What is MINT? . Jan 3 at 3:38
  • 2
    @AzorAhai-him- I just had to google it myself. MINT is German for STEM apparently.
    – adam.baker
    Jan 3 at 6:00
  • @adam.baker Thanks for clarifying it! Now corrected Jan 3 at 8:21

Your role as a supervisor is to teach your students to do research as best as they can, which means different things for different students. Some students will need to be trusted to handle things themselves, and some will need to get that code problem out of the way.

I think a general limit on what you would or wouldn't do to help a student (within professional boundaries) should only be limited by energy and time restraints. Refusing to do any particular thing on principle, without good reason, will create unnecessary insecurity between you and your student.

I have an hour set out every week where I meet my supervisors, where I can ask anything. They answer any questions they can and advice me on where to look further or how to do things myself if I ask about things that don't fit into that hour. This makes both sides have clear boundaries, I don't expect them to focus too much on me outside of that hour and they can expect me to bring up what I need from them during this meeting. I have the same setup when I am supervising students, and I also make an effort to ask questions and bring up relevant ideas to them when we have time left in our meeting after I have answered all of their questions.

I think the most important thing to focus on is the reason and result of the supervision; to help the student learn how to do things, and what a student needs will always be changing between different people and different situations. And clear boundaries from both directions is best for both of you since then you both know what to expect from each other.

  • Thanks - it's very helpful to understand the approach you take with your supervisors. Jan 4 at 6:14

There is no good answer to it.

I think, a supervisor should be present in the critical phases (e.g. experimental setup). Thereafter, I think one should be present enough to follow the research, to solve problems when the student gets stuck (e.g. coding problems or problems with instruments lead easily to unnecessary delays that do not provide a learning experience).

In quantitative terms: it all depends on the student, the project and the random issues we newer knew when starting the project.

Not the answer you're looking for? Browse other questions tagged .