An early-career lecturer takes on three master's students. Six months in, one of them sends a 60-page draft for the first time, a few weeks before the deadline, and the methods don't match the research question. The student says, “I thought you wanted me to finish it on my own first.” The lecturer thought, “If she isn't asking, she must be fine.” Neither meant any harm. They had simply never discussed how they would work together.
Supervising research is a distinct skill, different from classroom teaching and from doing research yourself. Many people learn it by repeating, or reversing, how they were supervised. This article offers a framework you can adapt.
The first meeting: agreeing expectations
Spend 45 to 60 minutes of the first meeting on questions people usually skip, and write the answers into a one-page document you both keep:
- Meeting rhythm: how often, in person or online, who books.
- Preparation: what the student sends and how far ahead, for example a one-page update 48 hours before.
- Drafts: sections or whole chapters, and how long feedback will take.
- Communication channels and reasonable response times for each.
- Who decides what: topic, methods, when it's ready to submit.
- Authorship if the work is published.
- When things go wrong: how much notice of delays, and who a neutral third person is if the relationship struggles.
This isn't a rigid contract. Review it each term and adjust.
Your role changes with the stage
| Stage | What the student most needs | Supervisor's role |
|---|---|---|
| Choosing a topic | Narrowing the question, judging feasibility | Point to gaps, warn when a topic is too broad |
| Literature review | Knowing what to read and when to stop | Suggest core sources, set a time limit |
| Design and methods | Choosing an appropriate approach | Check fit with the question, flag ethics |
| Data collection and analysis | Handling problems, staying motivated | Unblock technical issues, check progress |
| Writing | Structure and argument | Prioritised feedback |
| Finishing and examination | Confidence in answering questions | Mock defence, final review |
A common pattern is supervising too closely at the end, correcting sentences, and too loosely at the start, letting students choose an unworkable topic. Investing most heavily in the first month saves the most time later.
Feedback on drafts at three levels
Dense line-by-line comments on a draft with structural problems waste everyone's effort: the student polishes sentences in paragraphs that will be rewritten anyway. Give feedback in order:
- High level: is the question clear, does the argument flow, does the chapter structure work?
- Middle level: paragraphs, transitions, evidence for each claim.
- Low level: wording, citations, formatting.
If a draft still has high-level problems, comment only at that level, plus one or two examples of frequent low-level issues for the student to fix throughout. Open with two or three specific strengths rather than generic praise, then the three most important problems.
When a student gets stuck
Prolonged silence is usually a sign of being stuck, not of quiet progress. Common causes and responses:
- The task feels too big: break it into one-week tasks with concrete outputs.
- Perfectionism, reluctance to share rough work: set a routine of regular rough drafts and say explicitly that you want to see them.
- Technical blockage: connect them with someone in the group or a short course.
- Personal or health issues: listen, don't try to solve it for them, point to university support services, and explain extension procedures if needed.
Supervisors are not counsellors. Noticing the signs and referring students to the right service is the most important thing you can do.
The mock defence: the most valuable rehearsal
Two or three weeks before the examination, hold a mock with two or three colleagues or other students playing the committee. Ask them to pose the questions real examiners tend to ask: why this method, what is the biggest limitation, how far do the findings generalise? Afterwards, list with the student the answers that still wobbled and prepare them again. Students who have already heard the hard questions once are noticeably calmer on the day.
Authorship with students
Agree early, referring to ICMJE criteria or your field's norms. A common principle: a paper drawn mainly from a thesis is usually first-authored by the student, and the supervisor is an author if they contributed substantially to ideas, analysis or writing. Saying this at the first meeting prevents difficult conversations later.
Protecting your own boundaries
Supervising several students at once can swallow all your time. Some ways to keep it manageable: regular group meetings for shared issues, peer feedback among students before drafts reach you, and realistic turnaround times, such as ten working days for a chapter, instead of promising “right away”.
Try this: a meeting note template
Ask the student to send this within 24 hours, no more than half a page:
- Done since last meeting.
- Agreed in this meeting: decisions and directions.
- To do before next meeting, with deadlines.
- Open issues.
After a year, these notes form a progress log that is invaluable for progress reports, reference letters, and resolving any misunderstanding. For your next new student, draft a one-page expectations document covering the seven points above, send it before the first meeting, and use that meeting to revise it together.
Câu hỏi thường gặp
How often should a supervisor meet a thesis student?
It depends on the stage and degree level, but a regular rhythm such as every two to four weeks is common. Agree it at the start and ask the student to send a short update before each meeting.
How should I give feedback on a thesis draft?
Prioritise: address big issues with the question, argument and structure first, and detailed wording later. If a draft still has major problems, focus only on those.
What should I do if my student goes silent?
Reach out, since silence usually means being stuck. Break tasks down, encourage rough drafts, and refer the student to support services if personal issues are involved.
Should a supervisor be an author on papers from a student's thesis?
Yes, if they made a substantial contribution to ideas, analysis or writing under criteria such as ICMJE. The student is usually first author on papers mainly drawn from the thesis; agree this early.
How do I supervise several students without being overwhelmed?
Hold regular group meetings for common issues, have students give each other feedback before sending drafts to you, and set realistic turnaround times. Written meeting notes keep progress tracking light.
