A doctoral student arrives at her weekly meeting and opens with “I read a few more papers this week”. She summarises three articles for twenty minutes, her supervisor nods, suggests two more to read, and the meeting ends. The following week is the same. Three months later both feel nothing is moving, but neither says so. The meetings were not failing because the supervisor did not care. They were failing because there was never a question that needed deciding.
The supervisory relationship is built from hundreds of small interactions: meetings, emails, drafts sent, feedback received. You cannot control your supervisor's personality, but you can control much of how those interactions go.
Meetings: you own the agenda
Supervisors usually have several students and many other duties. Meetings work best when the student prepares the agenda. The day before, send a short email:
- Update: three lines on what you have done since last time, with outputs attached.
- Decisions needed: one or two specific questions, each with your proposed answer. For example: “I propose dropping the two schools with response rates below our threshold. Do you agree?”
- Feedback wanted: which part of which draft, and on what question.
- Next steps: what you will do before the next meeting.
A question with a proposal is far better than an open question. It shows you have thought it through, and your supervisor only needs to agree, adjust or explain why not.
After the meeting, send a three or four-line email the same day summarising what was agreed. It prevents misunderstandings and becomes a record of decisions.
Sending drafts that get read
| Do | Don't |
|---|---|
| Send a section readable in one sitting, with specific questions | Send a forty-page chapter with “could you take a look?” |
| State how finished it is: rough ideas or near-final | Leave them to guess, then get line edits on a rough draft |
| Ask in advance how long they need to read it | Send it on Friday evening hoping for comments by Monday morning |
| Send tracked changes for revisions | Send a new version without saying what changed |
Stating the level of polish matters a great deal. On an ideas draft you need comments on the argument, not the wording. Saying so upfront helps your supervisor read at the right level.
Emails that get answered
- A subject line that states the task: “Decision needed by Thursday: dropping two schools from the sample”.
- The question first, context afterwards.
- One topic per email. Three different questions in one email usually means only the first gets answered.
- If there is no reply after about a week, send a short, polite reminder. Busy supervisors are usually grateful for it.
When your supervisor is remote or very busy
Many doctoral students work with supervisors over video, or can only meet every few weeks. In that case, agree a fixed rhythm (a video call every two weeks at the same time, say), keep one shared document holding every agenda and decision, and find a co-supervisor or postdoc in the group for day-to-day questions.
When you disagree about direction
Disagreement is normal, and by your third or fourth year you may know the narrow corner of your topic better than your supervisor does. How to handle it:
- Understand their reasoning first: “Could you help me understand what concerns you about this direction?”
- Make your case with evidence, not preference: literature, pilot data, feasibility.
- Propose a test: a two-week trial, then decide.
- If you still disagree on something major, consider asking another committee member for a third view.
Remember that supervisors often know things not written anywhere: what examiners will ask, how journals will react, which tasks always take longer than planned.
Authorship and publications
Authorship is a common source of tension when it is not discussed early. Conventions differ between fields: in some the supervisor is usually last author, in others order follows contribution, in some it is alphabetical. Established authorship guidance, such as the ICMJE recommendations in biomedicine, is a useful reference point. Discuss author order before you start writing each paper, not when it is about to be submitted.
When the relationship is not working
Most difficulties can be improved by a frank conversation about expectations: how often you meet, how quickly feedback comes, how much independence you have. Prepare for that conversation with specific examples and specific proposals, not general complaints.
If problems are more serious, such as long periods with no supervision, disrespectful treatment or being asked to do inappropriate work, know your support channels: other committee members, the graduate programme director, and your institution's doctoral support office. Changing supervisor is a recognised procedure in many places. It is not easy, but sometimes it is the right decision for both sides.
Small things that build a good relationship
- Deliver what you promised, or warn in advance if you cannot.
- Let your supervisor know when their advice helped: “Your suggestion to recode that variable made the results much clearer.”
- Share good news: an accepted paper, a talk invitation, a travel grant.
- Propose ideas rather than waiting to be assigned tasks.
Before your next meeting, send the four-part agenda email above with at least one decision question and your proposed answer. Afterwards, send a three-line summary. Do this for four weeks in a row, then compare those meetings with the ones before.
Câu hỏi thường gặp
How should I prepare for a supervision meeting?
Send a short email beforehand with a progress update, decisions needed along with your proposed answers, the feedback you want and your next steps. Afterwards, send a summary of what was agreed.
What should I do if my supervisor is slow to give feedback?
Ask in advance how long they need, send manageable sections with specific questions and say how polished the draft is. If the agreed time passes, send a brief, polite reminder.
How do I handle disagreements with my PhD supervisor?
Understand their reasoning first, make your case with evidence and propose a small test to decide. If you still disagree on something major, seek a view from another committee member.
When should I discuss authorship with my supervisor?
Before you start writing each paper. Conventions vary across fields, and agreeing early prevents tension when the paper is ready to submit.
Can I change my PhD supervisor?
Many institutions have a procedure for it. Try a frank conversation about expectations first, and if the problem is serious, contact your programme director or doctoral support office.
