Interviews are one of the most valuable ways to gather qualitative data for a dissertation, but they need careful planning to produce credible results. This UK guide walks you through the types of research interview, how to design good questions, the ethics you must follow, and how to conduct and analyse interviews properly.
The Short Answer
A research interview study runs in a fixed order: define what you need to find out, choose an interview type that fits (usually semi-structured), write and pilot an interview guide, obtain ethical approval, recruit participants against clear criteria, gain informed consent in writing, record and transcribe, then analyse systematically — most often through thematic analysis. Two things sink more student projects than poor questions: starting to collect data before ethical approval, and treating transcription and analysis as an afterthought. Budget several hours of transcription and analysis for every hour of interview, and store the recordings on university-approved systems from the moment you make them.Types of Research Interview
There are three main types. Structured interviews use a fixed list of questions asked in the same order, giving consistency and easy comparison but little flexibility. Semi-structured interviews use a guide of key questions but allow you to follow up on interesting answers — this is the most popular choice for student research because it balances focus with depth. Unstructured interviews are open and conversational, rich in detail but harder to analyse. Choose the type that best fits your research questions.
Designing Your Interview Questions
Good questions are open, clear and neutral. Use open questions that invite explanation (“Can you tell me about…?”) rather than yes/no questions, and avoid leading questions that suggest an answer. Start with easy, general questions to build rapport before moving to more sensitive topics. Pilot your questions with one or two people first to check they are understood as you intended, and keep the number manageable so you can explore answers in depth.
Ethics and Informed Consent
Before you interview anyone, you almost always need ethical approval from your university. Participants must give informed consent, meaning they understand the purpose of your research, how their data will be used and stored, and that they can withdraw at any time. Provide a participant information sheet and a consent form, anonymise personal details, and store recordings securely in line with data-protection rules. Never begin data collection before your ethics approval is confirmed.
Want one-to-one support with your dissertation? Explore our dissertation writing service — trusted by UK students since 2001.
Conducting the Interview
Choose a quiet, comfortable setting or a reliable online platform, and test your recording equipment beforehand. Begin by restating the purpose and confirming consent, then work through your guide while listening actively and allowing silences that give the participant space to think. Use follow-up prompts such as “Can you say more about that?” to deepen answers. Stay neutral, avoid interrupting, and keep an eye on time so you cover your key questions.
Transcribing and Analysing Your Data
After each interview, transcribe the recording — writing out what was said — as soon as possible while it is fresh. You can then analyse the transcripts, most commonly through thematic analysis: coding the text, grouping codes into themes, and supporting each theme with representative quotations. Good analysis interprets what participants meant in relation to your research question, rather than simply summarising what they said.
Sampling and Recruitment: Who to Interview and How to Reach Them
Qualitative interviewing does not aim for a statistically representative sample; it aims for participants who can speak to your question with authority. That is purposive sampling, and it starts with written inclusion criteria — the role, experience, sector or characteristic that makes someone relevant, and anything that would exclude them. Writing the criteria down before you recruit protects you later, because you can explain in your methodology exactly who you spoke to and why.
Two variants are common in student work. Snowball sampling asks early participants to suggest others, which is useful for hard-to-reach groups but tends to produce a network with shared views — a limitation to acknowledge rather than hide. Convenience sampling uses whoever is accessible; it is defensible for a small project if you are candid about what it means for transferability.
Recruitment usually runs through gatekeepers: a manager, a course leader, a group administrator. Approach them with a short written summary of the study, your ethical approval reference and a participant information sheet, and let them circulate the invitation rather than contacting individuals directly, which avoids any sense of pressure. Expect a low response rate and over-recruit accordingly, because participants withdraw and diaries change.
Writing and Piloting an Interview Guide
An interview guide is not a questionnaire; it is a route map with room to move. The usual shape is a funnel. Open with easy, factual warm-up questions that build rapport and establish context. Move into the core questions organised around your research objectives, each one open and neutral. Close with a broad invitation — “Is there anything important we have not covered?” — which is where some of the most useful material appears.
Under each core question, write two or three probes in advance: “Can you give me an example?”, “What happened next?”, “How did that compare with…?”. Prepared probes are what separate an interview that produces stories from one that produces short answers, and they keep you from improvising leading questions under pressure.
Pilot the guide with one or two people who resemble your participants. A pilot tells you three things you cannot learn any other way: how long the interview actually takes, which questions are misunderstood, and which produce nothing. Revise the guide afterwards and say so in your methodology — piloting is a strength to report, not an embarrassment to conceal.
Online and Telephone Interviews
Remote interviews have become standard, and they change the practicalities more than the principles. Confirm consent in writing beforehand and again verbally on the recording, stating clearly that recording is about to begin. Use the platform your university licenses rather than a personal account, because institutional platforms keep the data inside approved systems and generate transcripts you are permitted to store.
Rapport takes more effort at a distance. Arrive early, allow a few minutes of conversation before you start, look at the camera rather than the screen when you can, and resist filling silences — a pause that feels long to you is often a participant thinking. On the telephone, where you lose visual cues, verbal acknowledgement matters more, and you should check periodically that you have understood.
Plan for failure. Agree in advance what happens if the connection drops, keep a second recording device as a backup, and note any interruption in your field notes because it becomes part of the context of that interview. Automatic transcripts from video platforms are a useful first draft but always need correcting against the audio, particularly for names and technical terms.
Consent, Data Protection and Storing Interview Data
Informed consent is a process, not a signature. Participants should receive an information sheet explaining the purpose, what participation involves, how long it will take, how their data will be used and stored, how they will be anonymised, and how to withdraw. The consent form should record separate agreement to take part and to be recorded, and you should keep the signed forms separately from the transcripts.
Interview recordings are personal data under UK data protection law, and your university’s approval process will set out the lawful basis and the conditions you must meet. In practice that means collecting only what you need, pseudonymising transcripts as early as possible, keeping the key that links pseudonyms to real names separate and secure, storing everything on institutional systems rather than personal cloud accounts or laptops, and deleting recordings on the timetable stated in your approved protocol.
Extra care applies where interviews touch on health, ethnicity, beliefs, sexuality or other sensitive categories, or where participants may be vulnerable. Those studies usually require a fuller ethics review and a plan for what you will do if a participant becomes distressed or discloses something you cannot keep confidential — a limit that must be stated on the information sheet before the interview, not afterwards. Your supervisor, ethics committee and university data protection officer are the authorities to consult, and their guidance overrides general advice.
Analysing Interview Data: Thematic Analysis Step by Step
Thematic analysis is the most widely used approach in student projects, and it follows six recognisable stages. First, familiarisation: read every transcript through, more than once, making notes rather than coding. Second, coding: label segments of text with short descriptive tags close to the participants’ own language. Third, generating themes: cluster related codes into candidate themes that say something about your research question.
Fourth, reviewing: test each theme against the coded extracts and against the full data set, splitting themes that contain two ideas and merging those that overlap. Fifth, defining and naming: write a sentence for each theme stating precisely what it claims. Sixth, writing up: build the findings around the themes, supported by extracts.
Practical points matter here. Keep a codebook recording each code, its definition and an example, because it is the audit trail that demonstrates rigour. A spreadsheet is perfectly adequate for a small project; dedicated software helps with larger data sets but does not analyse anything for you. And code the whole data set rather than mining it for quotations that fit an argument you have already decided on — that is the most common analytical failure in student dissertations.
Presenting Findings and Using Quotations Well
Findings chapters are organised by theme, not by participant, and each section should open with a claim of your own before the evidence appears. Introduce every quotation, present it, then interpret it — a quotation left to speak for itself does no analytical work, and long unbroken blocks of interview text read as padding.
Use pseudonyms or codes consistently, and include a short table of participants giving only the characteristics relevant to your analysis, such as role or years of experience. Remove anything that could identify a person or an organisation indirectly, including unusual job titles and small-town details, and say in your methodology what you removed and why.
Balance is an analytical decision. Show where participants disagreed as well as where they converged, indicate roughly how widely a view was held without implying false precision, and include a case that does not fit your theme where one exists. Interpreting rather than reporting is exactly the skill described in our guide to critical writing, and the findings chapter is where markers look for it.
Reflexivity, Bias and Demonstrating Trustworthiness
You are part of the instrument in interview research, so examiners expect you to account for your own influence. Reflexivity means stating your position in relation to the topic and participants — an insider interviewing colleagues gets different answers from an outsider — and considering how your questions, reactions and assumptions may have shaped the data.
Bias enters most often through leading questions and through selective attention during analysis. Reading your own transcripts as an interviewer is uncomfortable and instructive: you will find questions that suggested their answers, and noting them honestly in your limitations section is better than hoping the marker does not.
Trustworthiness is usually argued through four ideas: credibility, supported by careful coding and, where appropriate, checking interpretations with participants; transferability, supported by rich description of the context; dependability, supported by an audit trail of decisions; and confirmability, supported by grounding every claim in the data. Keeping a dated research diary from the first day is the cheapest way to be able to evidence all four, and it fits naturally alongside the structure set out in our guide to the chapters in a dissertation.
Frequently Asked Questions
How many interviews do I need for a dissertation?
There is no fixed number, but many undergraduate and master’s projects use between six and fifteen, aiming for the point where new interviews stop revealing new themes. Check your supervisor’s guidance.
Do I need ethical approval to interview people?
In almost all cases, yes. UK universities require ethical approval before you collect data from human participants. Never start interviewing before approval is granted.
What is the best type of interview for student research?
Semi-structured interviews are the most popular, because they keep your research focused while allowing you to explore interesting responses in more depth.
How long should a research interview last?
Semi-structured interviews in student projects commonly run between thirty and sixty minutes, and your pilot is the best guide to what your own guide requires. State the expected duration on the information sheet and keep to it, because overrunning erodes goodwill and produces tired answers. If a participant is willing to continue, ask rather than assume.
Do I need to transcribe every word?
It depends on your analytical approach. Thematic analysis usually needs a full verbatim transcript of what was said, though not the detailed notation of pauses and overlaps that conversation analysis requires. Automatic transcripts are a reasonable first draft but must be corrected against the audio, particularly for names, technical terms and negations.
Can I interview my own colleagues or fellow students?
Often yes, but declare it. Interviewing people you know raises questions about power, consent and candour, especially if you have any authority over them, and your ethics committee will want to see how you address that. Explain your relationship to participants in your methodology and reflect on how it may have shaped the answers you received.
How do I anonymise interview participants properly?
Replace names with pseudonyms or codes at the transcription stage, keep the list linking codes to real identities in a separate secure location, and remove indirect identifiers such as unusual job titles, organisation names and distinctive local details. Report only the characteristics your analysis needs, and describe in your methodology what you removed.
What happens if a participant withdraws after being interviewed?
Your information sheet should state a clear withdrawal deadline and what happens to data already collected. Up to that point, honour the request: delete the recording and remove the transcript from your analysis. After analysis has begun, complete removal may not be possible, which is precisely why the deadline must be explained before the interview rather than afterwards.