PhD supervision style: hands-on or hands-off?

There was a time when I doubted that meeting PhD students every week and getting involved in their projects was a good way of supervising and doing research. When I started supervising I met my PhD students once a week, but I was trying not to get too involved in their projects. Already by meeting them so often, I was feeling a bit guilty and unsure: "Is this the right thing to do? Will this affect their independence and their development as a researcher?"

A common type of supervisor that one finds in pure maths is the "hands-off" supervisor, they will typically say: "This is your project, not mine, do not come to me until you get results or get stuck; then, we discuss - and Latex everything before coming". Typically, students are a bit left on their own to struggle and try to find their way by themselves. Many times advisors give students projects and they do not work on them. The project of the student is only theirs: it is not the project of the supervisor too. There is no team spirit. This is in contrast to other disciplines: I know that in experimental Physics supervisors tend to be actively participating in the projects of their students. 

I think this "hands-off" style of supervision caused a great imprint on me and somehow I accepted, unquestioned, that this is the "(only) right way of supervising in pure maths". But when I tried to supervise like this, I felt worse: I lost track of the student - I did not know if they are stuck or having a hard time; I lost track of the project - so it was harder to give feedback; it became more energy consuming to get back to the project; I lost interest in the project. It feels cold; detached; bipolar: as if I should care, but not care at the same time.

A "hands-off" style of supervision does not fit who I am at all. But, for a long time, I did not truly believe that a "hands-on" style of supervision - based on weekly meetings and involvement from the supervisor - could work as well as the "hands-off" style. My view changed when I started meeting successful researchers in pure maths that had been supervised in the "hands-on" style.

A colleague of mine that benefited from the "hands-on" style also mentioned that even if the meeting with his supervisor was just to remember where they were at in the project, that was enough. There was no need to make progress at every meeting. That is a stress-free attitude: I think that, as a supervisor, I felt a lot of pressure to push projects forward at every meeting and to always end up with a "we achieved this" note.

In conclusion, I work best with a "hands-on" style. Of course, for every student and every supervisor, the style that works best is a different one. This may seem obvious now, but for a long time, I held many prejudices about mathematics and how research works. I have awakened from some of these prejudices over the years and found my own style as a researcher, supervisor, coach, mentor, teacher, and communicator.

We all need to find our own style and be authentic if we are to give our best.  

If you want to learn more about other styles of supervision, I recommend this article that I read today. Actually, some of the features in the 'coach-style' category fit me quite well:

The coach (‘steer my ambition’; ‘groom me into academics’): higher intensity, more personal and process-oriented

These supervisors are also very much involved in the growth of a candidate,[...]. They will put a lot of emphasis on styles of performance in public, scientific fora. They will stimulate candidates to go to presentation training courses and before examination they will suggest mock exams [...]. They try to understand the personality of the candidate and are aware of their personal circumstances. Whenever there are problems at home or with the (psychological) health of the candidate, the coach will try to be part of finding solutions. The coach is also interested in stimulating the scientific career of candidates beyond their PhD and will actively try to assist them in networking. In the first stages of PhD training, coaches are often involved in facilitation as well: with advice about time management, funding, [...].  - source: Effective PhD Supervision – Chapter Five – The Relationship between PhD Candidate and Supervisor by: Ahmed A. Wadee, Moyra Keane, Ton Dietz & Driekie Hay