Questions to Ask a Professor About Their Research
Questions for office hours, a first email meeting, or a conversation after a talk, covering what the professor is actually working on, how the field argues about it, and what a student would really do if they joined the group.
20 questions, each with the reason to ask it · includes a conversation guide
The questions
Open any question to see why it works.
- 1
How would you describe what you work on to someone outside your field?
This is a warm opener and a genuine test. Researchers who teach and write grants can do it in three sentences. If the answer immediately collapses into jargon, ask them to try again for a first-year student, which usually produces the clearer version.
- 2
What question are you trying to answer right now?
Websites and faculty pages describe a career, not a current project, and they are often years out of date. Asking about right now gets you the live work, which is the only work a new student could join.
- 3
How did you end up working on this?
The path is usually accidental: an adviser's project, a failed experiment, a dataset someone handed them. Hearing it tells you how much of a research career is chosen and how much is inherited, which is useful if you are trying to choose your own.
- 4
Which of your papers should I read first, and what should I take from it?
Asking for one paper rather than a list gets you the entry point instead of the greatest hits. The second half of the question matters, because the thing they want you to notice is often not the headline result.
- 5
What has surprised you in the last year?
Surprise is where the actual research is, as opposed to the parts already written up. This question often produces an unpublished result or a failed direction, which is far more informative than a summary of finished work.
- 6
What does a normal week look like for you?
Most people picture faculty life as thinking and teaching. The real answer usually includes grant writing, reviewing, committee work, and email, and the proportion tells you how much of their attention a student can realistically get.
- 7
Who disagrees with you, and what is their strongest point?
Every active field has open disputes, and the ability to state the other side fairly is a good proxy for how carefully someone works. It also gives you a map of the field: names, camps, and which questions are genuinely unsettled.
- 8
What result would make you abandon your current hypothesis?
You are asking for a falsification condition. A crisp answer means the project is designed to be tested. A long explanation of why the hypothesis is robust means the framework has become difficult to dislodge, which is worth knowing before you build three years of work on it.
- 9
Which of your findings has held up best, and which one do you worry about?
Almost nobody is asked the second half. Answers range from a replication that never worked to a sample that was too small, and the willingness to say so out loud tells you a great deal about how the group handles inconvenient data.
- 10
What is the hardest constraint right now: funding, data, equipment, or people?
The bottleneck shapes what students spend their time on. A group limited by data collection has students doing fieldwork or annotation. A group limited by funding may have students writing grant sections. Both are fine, but they are different jobs.
- 11
Who else in the world works on this, and how do you relate to them?
Small fields are networks, and knowing who collaborates with whom tells you where a student might visit, intern, or eventually apply. It also reveals whether this lab is central to the conversation or working slightly apart from it.
- 12
What could an undergraduate actually contribute here?
The honest range runs from real analysis to washing glassware and cleaning spreadsheets, and both exist in good labs. Ask what the first three months look like specifically, because a vague answer usually means there is no defined project waiting.
- 13
How do you decide what a new student works on?
Some advisers hand out a defined piece of a larger project, others expect students to arrive with a question. Neither is wrong, but a mismatch here is the most common reason a promising research relationship falls apart in the first year.
- 14
Who does the day-to-day supervision?
In most groups, the person a new student actually works with is a postdoc or a senior graduate student. Ask who that would be and whether you can talk to them, since they will affect your experience far more than the name on the door.
- 15
How does authorship work on your projects?
Norms differ sharply between fields, and inside a field between groups. Ask what a student typically has to do to be listed, and how order is decided. It is a completely normal question and it is much harder to ask after the work is done.
- 16
What should I learn before I would be useful, and how long does that usually take?
This turns a vague interest into a concrete next step: a language, a technique, a statistics course, a piece of software. The time estimate also tells you whether they are looking for someone who can commit a summer or several years.
- 17
What have your students gone on to do?
Where people land is a slow-moving signal about mentorship that no website reports. Listen for whether the professor knows the answer in detail. Advisers who stay in touch with former students usually invested in them while they were there.
- 18
What question in your field is nobody working on, and why not?
The why not is the interesting part: no funding, no method, no data, or a reputational cost to trying. This is one of the few questions that shows you the shape of a field from the inside rather than from its published output.
- 19
If funding were not a constraint, what would you work on instead?
Grant cycles quietly steer research toward what agencies will pay for. The gap between this answer and the current project tells you how much of the work is conviction and how much is available money.
- 20
What do people misunderstand about this work?
A good closing question, because it invites correction rather than summary. It often surfaces the overstated popular version of their results, which is exactly the version you probably arrived with.
Making an office hours conversation count
Practical guidance for the conversation itself.
Before you go
Before you go
- 1Read one paper, not five. Read the abstract, the figures, and the discussion, and write down two things you did not understand.
- 2Check the date on the faculty page against their recent publications. The page usually describes work from several years ago.
- 3Email with a specific ask and a time bound, such as fifteen minutes to ask about one paper. Open-ended requests are easy to leave unanswered.
- 4Bring a short note about what you can already do: a course, a technique, a language, a piece of software.
- 5Decide before you walk in whether you want to understand the work or join it, because those are two different conversations.
During the conversation
During the conversation
- Say plainly when you do not follow something. Researchers are used to explaining and it is far better than nodding through the answer.
- Ask about the failed version. What did not work is usually unpublished and is where most of the real explanation lives.
- Take notes on names and methods you do not recognise rather than stopping the flow each time.
- Do not pretend to have read something you have not. It becomes obvious in one follow-up question.
- Leave with one concrete next step: a paper, a person to email, or a date to come back.
If you are hoping to join the group
If you are hoping to join the group
- Ask to speak with a current student or postdoc alone. That conversation answers the mentorship questions the professor cannot answer about themselves.
- Find out how many people are in the group now and how many were there two years ago. Rapid growth and rapid shrinkage both change what your time looks like.
- Clarify whether the position is paid, for credit, or voluntary, and get the hours in plain numbers before you start.
- Ask what would make them tell you it is not working. It is easier to hear the exit conditions at the beginning.
- Send a short thank-you note with the one thing you are going to read next. It is the cheapest way to be remembered.