Quick answer: Good dissertation interview questions are open, neutral and traceable: each one exists because it helps answer your research question, not because it sounds interesting in isolation. Start from your research question and objectives, turn each objective into one or two open questions, add probes you can use if an answer is thin, order the questions so the conversation builds from easy to searching, and pilot the whole guide before your first real interview. A guide built that way produces data you can actually analyse; a list of questions written on the evening before fieldwork usually produces polite generalities that are hard to code, harder to quote, and impossible to connect back to the question you set out to answer.
Interview questions are also where methodology becomes visible. A marker can tell quickly whether questions were designed or improvised: designed questions use participants’ language, avoid leading them towards a hoped-for answer, and leave room for experiences you did not predict. Improvised questions bundle two questions into one, assume an experience the participant may not have had, or ask about abstract concepts — “empowerment”, “resilience” — that participants can only answer in equally abstract terms.
This guide takes you through the process in order: choosing the interview type, mapping questions to your research question, writing open questions, building probes, sequencing the guide, handling sensitive topics, and piloting. It assumes a semi-structured interview, the most common choice in UK dissertations, and notes where structured and unstructured approaches differ.
Choose the interview type before writing a single question
In a structured interview, every participant is asked the same questions in the same order, with little deviation. That consistency helps comparison but limits depth. In an unstructured interview, you begin with a broad topic and let the conversation develop, which can produce rich material but is harder to compare across participants and harder to justify to a marker who expects coverage of specific objectives.
A semi-structured interview sits between the two, and is the usual dissertation choice for good reason. You prepare a guide — questions and probes in a planned order — but may rephrase, reorder and follow up during the interview. The guide guarantees coverage of the topics your research question requires, while the flexibility lets you pursue an interesting answer and let participants use their own words. Name the type explicitly in your method and explain why it fits your question; markers notice when the choice is justified rather than assumed.
Your interview type also shapes how many questions you need. A semi-structured guide for a 45 to 60 minute interview typically contains around eight to twelve main questions, each with a couple of probes. That sounds few, but open questions generate long answers. A guide with twenty-five main questions will either overrun or force you to rush participants. Plan for depth, not coverage by volume.
Map every question to your research question
Before drafting, write out your research question and objectives in full, and keep them visible while you work. For each objective, ask: what would a participant need to tell me for this objective to be addressed? The answer to that question, turned into an invitation to talk, is your interview question. If you cannot say which objective a draft question serves, cut it or rewrite it — a question without a purpose is a question that wastes interview time and complicates analysis.
A simple mapping table keeps this honest: objectives down one side, draft questions beside them. An objective with no question beside it is a part of your study you will not be able to answer. An objective with six questions is probably two objectives disguised as one. A question serving no objective is either a warm-up question — legitimate, if labelled as such — or a curiosity belonging in a different study.
Consider a dissertation on how newly qualified nurses experience their first six months on a ward. Questions mapping to its objectives might include: “Can you walk me through your first week on the ward?”, “Who did you turn to when you were unsure about something, and what happened?”, and “Can you describe a moment when you felt confident in your role for the first time?” Each is open, concrete and tied to an objective — and each invites a story, which is where qualitative data lives.
Write open questions in participants’ language
Open questions cannot be answered with yes or no. They begin with how, what, tell me about, describe, or walk me through, and they invite narrative. “Tell me about a time when…” is often more productive than “What do you think about…?”, because it asks for a specific experience rather than a general opinion — and experiences give you incidents and language you can analyse, where opinions alone often give you slogans.
Keep each question to one idea, in plain language. A double-barrelled question — “How did your manager and your colleagues support you, and did that change over time?” — forces the participant to choose which half to answer, and you will rarely get both. Split it. The same applies to jargon: participants may not use terms like “professional identity” as your literature does. Ask about what happened and how it felt, and let concepts emerge in analysis rather than imposing them in the question.
Above all, keep questions neutral. A leading question smuggles your hypothesis into the wording: “How frustrating did you find the lack of support?” assumes frustration and a lack of support, and a polite participant will often agree rather than correct you. The neutral version — “How would you describe the support you received?” — lets the participant set the tone. Neutrality also protects your analysis: findings that merely echo your questions are easy for a marker to spot and hard to defend.
Build probes, and know when to use them
Probes are the short follow-ups that turn a thin answer into usable data, and they should be written into your guide in advance. The most useful are simple: “Can you tell me more about that?”, “Can you give me an example?”, “What happened next?”, and a deliberate silence after an answer, which often produces the most reflective material of all. Write two or three beside each main question.
Probes also rescue interviews that stall. If a participant gives a one-line answer, a probe signals that detail is welcome. If they drift off topic, a gentle steering probe — “That’s helpful; coming back to your first weeks, …” — returns the conversation without shutting them down. And if they say something unexpected and important, a probe lets you follow it: the guide is a plan, not a script.
One caution: probes should clarify, not interrogate. Pressing for detail on a topic a participant is clearly uncomfortable with damages rapport and can cross an ethical line. Decide in advance how you will handle a participant who does not want to go further — usually by acknowledging it, offering to move on, and returning only if they reopen it.
Sequence the guide like a conversation
Order matters as much as wording. Start with warm-up questions that are easy and factual: the participant’s role, how long they have been in it, a typical day. These build rapport, settle nerves, and give context that makes later answers intelligible. A participant who has just described their daily work will answer a question about a difficult incident far more fully than one asked about it cold.
Place your core questions in the middle, once rapport is established and before fatigue sets in, with sensitive questions towards the later part of that section. End with closing questions that give the participant the last word: “Is there anything else about this experience that you think I should understand?” Closing questions frequently produce the richest material, because the participant can now see the whole conversation and add what it missed. Always close by thanking them, explaining what happens next with their data, and checking how they feel about what they have shared.
Handle sensitive topics and ethics in the questions themselves
If your topic touches on distress, failure or personal health, your questions need ethical design, not just ethical approval. Avoid wording that implies blame: “Why did you not ask for help?” accuses, where “What made asking for help easy or difficult?” invites explanation. Offer participants control explicitly — in the introduction and again before a sensitive section — by saying they can skip, pause, or stop, and mean it.
Your information sheet and consent process should match the guide: participants should know the topics in advance, roughly how long the interview will take, and that they can withdraw. If a question emerged from your literature review, you should be able to explain that link in your methodology chapter. That traceability — from literature gap, to objective, to question, to data — is what makes a qualitative methodology chapter convincing.
Pilot the guide, then revise it properly
Never take an untested guide into your first real interview. Pilot it with one or two people who resemble your participants but will not be part of your sample, and time it: if it runs to ninety minutes when you planned sixty, cut questions rather than rushing real participants.
Piloting reveals four kinds of problem: questions participants misunderstand (reword them), questions producing thin or generic answers (reframe towards concrete experience), questions leading nowhere (cut them), and gaps — topics participants raise that your guide never asks about (add questions or probes). After each pilot, revise the guide and note what changed and why; that revision history is useful evidence in your methodology chapter that the instrument was tested, not merely written.
Two practical checks complete the pilot. First, record and listen back, even briefly: you will hear your own habits — interrupting, finishing sentences, asking a second question before the first is answered — more clearly than any advice can convey. Second, check that your opening script feels like a conversation starting, not a form being read aloud. Participants relax when you sound prepared rather than rehearsed.
Questions to avoid, and what to write instead
Some question types cause trouble so reliably that they are worth naming. Closed questions — “Did you enjoy your placement?” — produce yes or no, and belong in surveys, not interviews. Double-barrelled questions split the participant’s attention. Leading questions plant your answer. Hypothetical questions — “What would you do if…?” — produce speculation rather than experience, unless speculation is genuinely your topic. And jargon-heavy questions produce jargon-heavy answers that tell you what participants think they should say, not what happened.
For each type, the fix is usually the same: return to experience. Instead of “Did you feel supported?”, ask “Can you describe a time when you needed support, and what happened?” Instead of a combined question about professional development and mentoring, split it and ground it in a specific experience. The pattern is consistent: one idea, plain words, a concrete experience, and room for the participant to answer in their own terms.
Frequently asked questions
How many interview questions should a dissertation guide contain?
For a semi-structured interview of 45 to 60 minutes, around eight to twelve main questions, each with a few probes, is a workable range. Open questions generate long answers, so a longer list usually means rushing participants or overrunning. Let the pilot tell you whether the count is right for your topic and your participants.
Should every interview question link to my research question?
Every main question should, yes — that is the test for keeping it in the guide. Warm-up questions are the exception: they build rapport and context rather than answering an objective directly, and they should be recognisable as such. If a question serves neither an objective nor rapport, it probably does not belong in the guide.
What is the difference between a question and a probe?
A question opens a topic; a probe deepens an answer already given. “Can you describe your first week on the ward?” is a question. “Can you give me an example of that?” or “What happened next?” are probes. Writing probes into the guide in advance means you are not searching for follow-ups while also listening, noting and managing the conversation.
Do I need to pilot if my supervisor has already approved the guide?
Yes. Supervisor approval means the guide looks right on paper; piloting shows whether it works in conversation. Questions that read clearly can confuse participants when spoken, and only a pilot reveals that — along with timing, gaps and questions that produce thin answers. Note the changes you make after piloting in your methodology chapter.
How do I ask about sensitive topics without causing distress?
Use neutral, experience-based wording, warn participants before a sensitive section, and give them explicit control to skip, pause or stop. Avoid wording that implies blame, and do not press for detail if a participant signals discomfort. Your ethics approval, information sheet and consent process should all reflect the same approach as the questions themselves.
Can I change the guide during fieldwork?
In a semi-structured design, minor changes are normal: rewording a question that consistently confuses participants, or adding a probe for a theme that keeps emerging. Record any changes and the reason for them, and discuss substantial changes with your supervisor, since they affect comparability across interviews. A guide that evolves transparently is a strength; a guide that changes silently is a methodology problem.
A strong interview guide is not a long list of clever questions. It is a short, purposeful set of open invitations, mapped to your research question, ordered like a conversation, and tested before it matters. Build it that way, pilot it honestly, and your interviews will give you the kind of detailed, quotable, analysable material that makes the rest of the dissertation — coding, analysis, discussion — noticeably easier.