Face-to-face Semi-structured Interview
- Introduced by
- Rooted in sociological interview traditions; systematised by researchers including Robert Merton and Paul Lazarsfeld (focused interview, 1940s) and later elaborated by Steinar Kvale
- Year
- 1940s onward; widely codified in the 1980s–1990s
- Type
- Qualitative data collection technique
- Family
- Data collection
What does it do?
A face-to-face semi-structured interview is a qualitative data collection technique in which a researcher meets a participant in person and follows a prepared topic guide of open-ended questions while retaining the flexibility to probe, reorder, and explore emerging themes.
It combines the consistency of a predetermined agenda with the depth and responsiveness of an open dialogue, making it one of the most widely used methods in qualitative and mixed-methods research across the social, health, and educational sciences.
The intuition behind the method
Think of it as a guided conversation: the researcher arrives with a map of topics to cover, but the route through them is negotiated in real time. Unlike a fully structured interview where every question is fixed, the semi-structured format allows the researcher to follow up on unexpected insights, ask for elaboration, or adjust the order of questions based on what the participant says. The physical presence of both parties supports rapport and allows the researcher to read non-verbal cues — a dimension unavailable in phone or online alternatives.
History
Semi-structured interviewing has roots in the focused interview developed by Robert Merton, Marjorie Fiske, and Patricia Kendall in the 1940s to explore the subjective responses of media and wartime communication audiences. The technique was elaborated through qualitative traditions in sociology and anthropology across the mid-twentieth century. Steinar Kvale's 1996 handbook InterViews brought systematic methodological guidance to the practice and became a foundational reference. The explicit distinction between structured, semi-structured, and unstructured interview formats was consolidated in research methods textbooks by the 1980s and is now standard in social science training.
When to use it
Use a face-to-face semi-structured interview when the research question requires in-depth, contextual understanding of participants' views, experiences, or meanings — particularly for sensitive, complex, or exploratory topics where spontaneous follow-up is valuable. The method is well suited to health research, education studies, organisational research, and social inquiry. It is the preferred modality when building rapport is important, when participants have limited digital literacy, or when non-verbal cues are analytically relevant. Do NOT use it when a large, representative sample is needed (a survey is more appropriate), when the topic is simple enough for a questionnaire, when geographic or time constraints make in-person contact impractical (use online or telephone alternatives), or when the research requires strictly comparable data across many respondents (use a structured interview).
Limitations and considerations
- Resource-intensive: travel, scheduling, transcription, and analysis demands make it costly in time and money.
- Findings are not statistically generalisable; conclusions describe patterns within the sample, not population parameters.
- Interviewer effects — the researcher's presence, identity, and manner — can influence participant responses.
- Geographical constraints restrict the achievable sample to participants who can meet in person.
Source conditions do not replace independent scientific review or validation by an analysis engine.
Strengths
- Physical co-presence enables rapport-building and access to non-verbal cues unavailable in remote methods.
- The flexible guide allows the researcher to pursue unexpected but relevant themes that emerge mid-interview.
- Produces rich, contextualised data that reveals the complexity of participants' perspectives and meanings.
- Widely accepted across disciplines; extensive methodological literature supports rigour and reporting standards.
- Participant engagement tends to be higher in person than in self-completion or telephone formats.
Applications
- Exploring patients' experiences of diagnosis, treatment, or health service use in health research.
- Investigating teachers' and students' perceptions of educational practices, policies, or learning environments.
- Examining employees' experiences of organisational change, leadership, or workplace culture.
- Understanding the decision-making processes of policy-makers, practitioners, or service users.
- Gathering expert knowledge in fields where experience and tacit understanding matter as much as formal evidence.
- Conducting needs assessments or programme evaluations where stakeholder perspectives are central.
How it works
How it works
The researcher develops an interview guide — a list of open-ended questions and sub-questions organised around the key research themes. The guide is piloted with one or two test participants to check clarity, flow, and time requirements. It serves as a flexible agenda rather than a script: the researcher may rephrase questions, probe further, or skip items that have already been addressed naturally in the conversation.
Participants are selected purposively — chosen because they have direct experience of the phenomenon under study. Sample sizes in semi-structured interview studies typically range from 8 to 30, depending on the scope of the study and the criterion of thematic saturation. Ethical requirements (informed consent, confidentiality assurance, right to withdraw) are addressed before the interview takes place.
The researcher meets the participant at an agreed location. The conversation is guided by the interview topic guide but remains responsive to what the participant raises. Probing techniques — such as asking 'Can you say more about that?' or 'What did you mean by…?' — encourage depth and elaboration. The researcher attends to verbal content and non-verbal signals (pauses, discomfort, enthusiasm) to guide the pace and direction of the exchange.
With the participant's consent, the interview is audio-recorded. The recording is then transcribed verbatim or in an edited form appropriate to the analytic approach. The researcher typically also writes brief field notes capturing contextual impressions — setting, tone, notable non-verbal moments — immediately after the session.
Transcripts are analyzed using a qualitative method appropriate to the research question — most commonly thematic analysis, interpretive phenomenological analysis, grounded theory coding, or content analysis. Patterns, themes, and categories are identified across participants to produce findings that address the research aim.
Common pitfalls
- Using the guide as a rigid script — treating open-ended questions as a fixed sequence prevents the responsive probing that defines the method and reduces it to a structured interview in practice.
- Insufficient probing — accepting brief or vague answers without follow-up yields thin data; probes such as 'Can you give me an example?' are essential.
- Neglecting reflexivity — failing to acknowledge and reflect on how the researcher's identity, assumptions, or rapport with the participant may have shaped the data weakens the credibility of findings.
- Inadequate recording or transcription — poor audio quality or selective transcription introduces data loss that undermines subsequent analysis.
- Overstating saturation — declaring thematic saturation with too few interviews (typically fewer than 8–10 for a focused study) risks missing important variation in participants' experiences.
Frequently asked questions
How is a semi-structured interview different from an unstructured interview?
A semi-structured interview follows a prepared topic guide of key questions, ensuring that the same broad themes are addressed across all participants. An unstructured interview begins with a single opening question or topic and proceeds entirely through the participant's own narrative, with the researcher intervening minimally. Semi-structured interviews are more consistent across the sample; unstructured interviews allow greater narrative freedom but are harder to compare.
How many participants do I need?
There is no single rule, but 8–20 participants is a commonly cited range for a focused qualitative study using semi-structured interviews. The guiding criterion is thematic saturation — the point at which additional interviews consistently produce no new themes. With fewer than eight participants, saturation is difficult to achieve unless the study is highly focused or uses a purposive sample of rare experts.
What makes the face-to-face modality worth the extra effort compared with online or telephone alternatives?
In-person interviews support rapport development, allow the researcher to notice and respond to non-verbal cues (hesitation, discomfort, enthusiasm), and tend to result in longer, more reflective responses. They are especially valuable for sensitive topics or with participants who are less comfortable with digital platforms. When topic sensitivity or participant characteristics make these advantages important, the additional logistical cost is usually justified.
Do I need to transcribe the interviews verbatim?
It depends on the analytic approach. For discourse analysis or conversational analysis, verbatim transcription including pauses and overlaps is essential. For thematic analysis or interpretive phenomenological analysis, near-verbatim transcription with light editing for readability is standard. The key principle is that the transcription level should match the grain of the analysis; transcribing more than the analysis requires wastes time, while transcribing less risks data loss.
How do I handle sensitive topics in a face-to-face interview?
Sensitive topics require careful attention to the setting (a private, comfortable space), explicit informed consent, clear communication of confidentiality and the right to withdraw at any time, and a slow build of rapport before probing difficult areas. The researcher should also be prepared with signposting information for support services if participants become distressed, and should debrief participants at the close of the interview.
Related methods
Sources
Bryman, A. (2016). Social Research Methods (5th ed.). Oxford University Press.
ISBN: 9780198722519Copied.Kvale, S., & Brinkmann, S. (2009). InterViews: Learning the Craft of Qualitative Research Interviewing (2nd ed.). Sage.
ISBN: 9780761925422Copied.