Interviews in research are a primary, qualitative data-collection method in which a researcher asks a participant a planned set of questions to elicit detailed, first-person accounts of their experiences, beliefs and behaviour. Use interviews when you need rich, in-depth understanding of how and why people think and act, rather than how many people hold a view. They suit exploratory studies, sensitive topics, and questions where the participant’s own words and meanings matter more than countable responses.
The main types of interviews are structured, semi-structured and unstructured, sitting on a spectrum from fixed questions to a free-flowing conversation; in-depth and group (focus-group) interviews are common variants. The right choice depends on how much you already know about the topic and how much flexibility your research questions demand.
What is an interview in research?
An interview is a purposeful, professional conversation in which a researcher (the interviewer) asks questions and a participant (the interviewee) answers, with the exchange recorded for later analysis. It is one of the most widely used methods of data collection in qualitative and mixed-methods research because it gives the researcher direct access to a participant’s perspective, in their own words. Unlike a questionnaire, an interview is interactive: the researcher can probe, clarify, follow an unexpected lead, and check that a question has been understood. For a fuller map of where interviewing sits among other techniques, see our guide to the methods of data collection.
Interviewing rests on an interpretivist assumption: that reality is constructed through the meanings people attach to their experiences, and that those meanings are best uncovered by talking with people in depth. This is why interviews dominate qualitative designs such as phenomenology, grounded theory, case study and narrative inquiry. They can, however, also serve quantitative or mixed designs — a tightly structured interview is essentially an orally administered survey.
When to use interviews (and when not to)
Choose interviews when your research question is exploratory, when the topic is complex, sensitive or personal, or when you need to understand processes, reasoning and context rather than measure frequencies. They are ideal early in a project to scope a poorly understood area, or to explain surprising survey results.
Interviews are a strong choice when:
- You need rich, detailed accounts of lived experience, motivation or decision-making.
- The topic is sensitive (e.g. illness, bereavement, workplace conflict) and rapport matters.
- Your sample is small, specialised or hard to reach (e.g. senior clinicians, policy experts).
- You are exploring a new area where you cannot yet write fixed, closed questions.
- You want participants to raise issues you had not anticipated.
Interviews are usually the wrong tool when you need statistically generalisable findings from a large sample, when you require strictly standardised measurement, or when anonymity is essential and a face-to-face encounter would deter honesty. In those cases a questionnaire or large-scale survey is more appropriate.
Types of interviews
The classic typology places interviews on a continuum of structure. At one end the interviewer controls everything; at the other, the participant largely leads. Three core types — structured, semi-structured and unstructured — are joined by two common formats defined by depth and group size: the in-depth interview and the group or focus-group interview.
1. Structured interviews
Every participant is asked exactly the same questions, in the same order, with the same wording, and often with fixed response options. Structured interviews maximise comparability and reliability and are quick to analyse, but they suppress flexibility: the interviewer cannot probe or pursue an unexpected answer. They suit large samples and topics that are already well understood.
2. Semi-structured interviews
The most popular type in student and professional qualitative research. The interviewer works from an interview guide — a list of themes and core questions — but is free to change the order, reword questions, and probe deeper as the conversation unfolds. This balances focus (so interviews remain comparable) with the freedom to capture rich, unanticipated detail. If you are unsure which type to use, the semi-structured interview is usually the safest, most defensible default.
3. Unstructured interviews
Closer to a guided conversation. The researcher may begin with a single broad prompt (“Tell me about your experience of returning to study as a mature student”) and then follow the participant’s lead, probing as themes emerge. Unstructured interviews yield the deepest, most participant-driven data but are demanding to conduct, hard to compare across participants, and time-consuming to analyse. They are common in ethnography and life-history research.
4. In-depth interviews
An in-depth interview is a one-to-one, usually semi- or unstructured encounter aimed at exploring a single participant’s perspective in great detail, often over 60–90 minutes. The term describes the intent (depth) rather than a fixed level of structure, and it overlaps heavily with the semi-structured format.
5. Group and focus-group interviews
Here the researcher interviews several participants together. A focus group (typically 6–10 people led by a moderator) deliberately uses interaction between participants — agreement, disagreement, building on one another’s ideas — as data. Group interviews are efficient and surface a range of views quickly, but they risk groupthink, dominant voices and reduced confidentiality, so they suit non-sensitive topics. For a deeper treatment of moderation, recruitment and analysis, see our dedicated guide to focus groups.
Comparing the main interview types
| Interview type | Structure | Flexibility | Best for |
|---|---|---|---|
| Structured | Fixed questions, fixed order and wording | Low | Large samples; comparability; well-known topics; mixed-methods surveys |
| Semi-structured | Guide of core questions; order and wording flexible | Medium | Most qualitative dissertations; balancing focus with rich detail |
| Unstructured | One or two open prompts; participant leads | High | Ethnography; life histories; exploring unknown territory |
| In-depth | Usually semi/unstructured, one-to-one, extended | Medium–High | Deep single-case perspectives; sensitive personal experience |
| Group / focus group | Moderated discussion among 6–10 people | Medium | Surfacing shared views; reactions to a product, policy or idea |
“The qualitative research interview attempts to understand the world from the subjects’ points of view, to unfold the meaning of their experiences.” (Source: Kvale & Brinkmann, 2009)
How to design an interview guide
The interview guide is the backbone of a semi-structured or in-depth interview. It is not a rigid script; it is a structured prompt sheet of themes, core questions and possible probes, derived directly from your research questions and literature review. A good guide keeps the conversation on track while leaving room to follow the participant.
Follow these steps to build a guide:
- Anchor it in your research questions. Every theme in the guide should map to a research question; if it does not, cut it.
- Write open questions. Begin questions with how, what or tell me about so participants cannot answer with a bare yes or no.
- Funnel from broad to specific. Open each theme with an easy, general opening question, then narrow towards detail and sensitive points — the funnelling technique.
- Plan probes. Note follow-ups such as ‘Can you give me an example?’, ‘What did that feel like?’ or ‘Why do you think that happened?’ under each main question.
- Sequence sensibly. Start with warm-up questions to build rapport, place sensitive material in the middle, and end on a positive, reflective note.
- Pilot the guide. Test it with one or two people, then refine wording and order before collecting real data.
Avoiding leading and double-barrelled questions
A leading question signals the answer you expect (‘Don’t you agree the new policy improved morale?’) and biases the response. Rephrase neutrally: ‘How, if at all, did the new policy affect morale?’ Avoid double-barrelled questions (which ask two things at once), jargon, and loaded wording. The cleaner your questions, the more your data reflect the participant rather than the interviewer — a core concern of reliability and validity in qualitative work.
Worked example: from interview guide to coded data
Interview-guide excerpt (one theme):
- Opening (broad): “Tell me about your first few weeks at university.”
- Core question: “How did you find managing your own study time?”
- Planned probes: “Can you give me a specific example?” / “What did that feel like?” / “What helped, or got in the way?”
How a response is probed:
Participant (P7): “Honestly, the first month was a shock. No one told me when to do anything.”
Interviewer (probe): “A shock — can you give me a specific example of when you felt that?”
P7: “I had three deadlines in one week and I’d done nothing, because at school the teachers chased you. Here, if you don’t organise yourself, you just sink.”
How the response is later coded (thematic analysis):
- Initial codes: “lack of external structure”, “deadline clustering”, “sink-or-swim”, “loss of teacher monitoring”.
- These codes recur across P3, P7 and P11 and are grouped into a candidate theme: “From external to self-imposed structure”.
- The theme is refined against the data set, named, and reported with this quotation as illustrative evidence.
This is a complete loop: a funnelled, open question → a real-time probe for a concrete example → codes → a theme. For the full coding procedure, see thematic analysis.
Interview modes: face-to-face, phone and video
Interviews can be conducted in person, by telephone, or over video platforms such as Zoom or Microsoft Teams. Each mode trades richness against reach and cost.
- Face-to-face: richest rapport and access to body language; best for sensitive topics. Costlier and limited by geography.
- Telephone: cheap, flexible and good for reaching dispersed participants; loses visual cues and can feel less personal.
- Video: a strong compromise — visual cues plus global reach and easy recording — but depends on connectivity and digital confidence, and may exclude some participants.
Conducting the interview: rapport, probing and listening
The quality of interview data depends heavily on how the interview is conducted. Three skills matter most.
- Building rapport: open with easy, non-threatening questions; explain the purpose and confidentiality; be warm but neutral. Participants disclose more when they feel safe and unjudged.
- Probing: use neutral follow-ups (‘Tell me more’, ‘What happened next?’, ‘Why was that important to you?’) to deepen thin answers without leading the participant.
- Active listening: give the participant space, tolerate silence, and resist the urge to fill every pause — silence often produces the most candid material. Listen more than you speak; a good rule of thumb is that the participant should talk far more than you do.
Recording, transcription and analysis
With informed consent, audio-record the interview so you can concentrate on listening rather than note-taking. The standard analytic pipeline is record → transcribe → code → theme.
- Record with a reliable device (and a backup), having obtained written consent and explained how data will be stored.
- Transcribe verbatim — usually intelligent verbatim, which keeps meaning while tidying filler. Transcription is slow (often 4–6 hours per hour of audio); our practical tips to transcribe an interview can speed this up.
- Familiarise and code by reading transcripts closely and tagging meaningful segments with codes.
- Develop themes by grouping related codes, then review, define and name them.
The dominant approach to analysing interview data is thematic analysis, systematised by Braun and Clarke (2006) as a six-phase process: familiarisation, generating initial codes, searching for themes, reviewing themes, defining and naming themes, and producing the report. Other options include framework analysis, grounded theory coding and, for language-focused questions, content analysis.
Strengths and limitations of interviews
Strengths:
- Produce rich, detailed, first-person data with context and nuance.
- Flexible: the interviewer can clarify, probe and follow unexpected leads.
- High response quality and the chance to access meaning, not just behaviour.
- Suit sensitive topics and hard-to-reach or specialised participants.
Limitations:
- Time and cost: recruiting, conducting, transcribing and analysing interviews is labour-intensive.
- Interviewer bias: the researcher’s wording, tone and reactions can shape answers; participants may also give socially desirable responses.
- Limited generalisability: small, purposive samples mean findings are not statistically representative of a wider population.
- Skill-dependent: data quality hinges on the interviewer’s questioning and listening skills.
These trade-offs are why many designs combine interviews with quantitative methods; our overview of quantitative vs qualitative research sets out how the two complement each other.
Common mistakes to avoid
- Asking leading or loaded questions that telegraph the ‘right’ answer.
- Talking too much. Over-explaining or interrupting steals the participant’s floor.
- Sticking rigidly to the guide in a semi-structured interview and missing rich tangents.
- Failing to probe, so answers stay shallow and generic.
- Neglecting consent and confidentiality, or skipping ethical approval.
- Under-planning analysis: collecting hours of audio with no clear coding strategy and drowning in data.
- Over-claiming generalisability from a small qualitative sample.
How to do interviews well
Strong interview studies are planned end to end. Align the interview type with your research questions, build a piloted guide of open, funnelled questions, secure ethical approval and informed consent, and rehearse your probes. During the interview, build rapport, listen actively and let silence work. Afterwards, transcribe promptly, code systematically, and report themes with verbatim quotations as evidence. Document every decision so your methodology chapter is transparent and defensible.
Struggling to design or write up your interviews?
Our academics can help you plan your interview study, build a robust guide and write a watertight methodology chapter.
