Skip to content
Back to search results

In-Depth Interview — In-Depth Qualitative Interview

Introduced by
Rooted in sociological interviewing traditions; systematised by researchers including Steinar Kvale and Herbert J. Rubin
Year
Mid-20th century (formalised in qualitative social research from the 1950s onward)
Type
Qualitative research method
Family
Interview Methods
Font size

What does it do?

The in-depth interview is a one-to-one qualitative data-collection method in which a researcher engages a participant in an extended, open-ended conversation to elicit rich, detailed accounts of experiences, perceptions, beliefs, or meanings.

Unlike structured surveys, the interview guide serves as a flexible road map rather than a fixed script, allowing the researcher to probe unexpected directions as they emerge. The approach is foundational to qualitative inquiry and is used directly as a primary method or as the data-collection arm of phenomenology, grounded theory, narrative analysis, and other frameworks.

The intuition behind the method

Think of an in-depth interview as a purposeful, intellectually guided conversation. A researcher meets a participant — say, a nurse describing how she navigates end-of-life care decisions — and instead of ticking boxes on a form, they listen carefully, ask follow-up questions, and invite the participant to elaborate, correct, and deepen what she has said. The conversation might last 45 to 90 minutes. The researcher records and later transcribes every word, then works through the transcript systematically to extract meaning. The power of the method lies in this depth: a participant can say things a questionnaire would never think to ask, and the researcher can pursue those threads in real time.

History

Systematic use of extended individual interviews as a social-science method dates to the Chicago School sociologists of the 1920s and 1930s, who employed life-history interviews alongside ethnographic fieldwork. The method was formalised and theorised most influentially by Steinar Kvale, whose 1996 book InterViews provided a comprehensive epistemological and practical framework that remains a standard reference. Herbert and Irene Rubin's Qualitative Interviewing (1995; 2nd ed. 2005) further elaborated the craft dimensions of the method. The rise of interpretive paradigms in the social sciences from the 1970s onward, and the subsequent development of analytic frameworks such as interpretive phenomenological analysis (Smith, 1996) and constructivist grounded theory (Charmaz, 2006), consolidated the in-depth interview as the primary data-collection tool in much qualitative health, education, and social research.

When to use it

In-depth interviewing is appropriate when the research question requires understanding how individuals experience, perceive, or make meaning of a phenomenon from their own perspective. It is the method of choice when the topic is sensitive, complex, or insufficiently understood for a survey instrument to be adequate, and when individual variation in experience is as important as commonalities. It suits exploratory, descriptive, and interpretive purposes across social sciences, health, education, and organisational research. The method is NOT appropriate when the goal is to estimate population frequencies or test causal hypotheses — a survey or experiment is more suitable. It is also unsuitable when participants cannot reflect on and articulate their experience verbally, or when time and resources do not permit the transcription and careful analysis that the method demands.

Limitations and considerations

  • Findings are not statistically generalizable; the sample is purposive rather than representative, so population-level inference is inappropriate.
  • Highly resource-intensive: a single study may involve 20 or more hours of recording, many more hours of transcription, and extensive iterative analysis.
  • Data quality depends heavily on the researcher's interviewing skill — rapport-building, active listening, and the ability to probe without leading.
  • Participant self-report is subject to recall bias, social desirability effects, and the limits of verbal articulation.
  • Saturation is difficult to judge precisely; there is no formula, and under-sampling risks missing important variation while over-sampling wastes resources.

Source conditions do not replace independent scientific review or validation by an analysis engine.

Strengths

  • Produces rich, detailed, first-person accounts that reveal depth and nuance unavailable to surveys or structured observations.
  • Flexible and responsive: the researcher can pursue unexpected directions, clarify ambiguities, and adapt questions as understanding develops.
  • Well-suited to sensitive or stigmatised topics where one-to-one privacy encourages candour that group settings might inhibit.
  • Applicable across a wide range of disciplines and research questions without requiring specialist laboratory equipment.
  • Captures the participant's own language, concepts, and interpretive frames rather than imposing the researcher's categories in advance.

Applications

  • Exploring how patients understand and manage chronic illness, treatment decisions, or interactions with healthcare professionals.
  • Investigating organisational experiences — how employees navigate workplace change, leadership, or professional identity.
  • Understanding how marginalised or hard-to-reach populations experience social institutions, policy, or discrimination.
  • Examining decision-making processes among professionals (clinicians, teachers, policymakers) in naturalistic context.
  • Generating the rich participant accounts needed as primary data for phenomenological, grounded-theory, or narrative research designs.
  • Piloting themes before designing a large-scale survey instrument, to ensure the questionnaire captures the concepts that actually matter to participants.

How it works

How it works

1

The researcher formulates the research question and constructs a semi-structured (or unstructured) interview guide: a set of open-ended topic areas and probe questions designed to elicit relevant narratives. The guide is piloted with one or two participants not in the final sample and revised based on what the questions actually produce. Ethical approval, informed consent procedures, and recording permissions are secured at this stage.

2

Participants are selected purposively — chosen because they have direct experience of the phenomenon under study rather than because they are statistically representative of a population. Snowball sampling is common when the target group is hard to reach. Sample size is determined by theoretical (or informational) saturation: recruitment continues until new interviews yield no new themes or conceptual categories, a point typically reached between 10 and 30 participants depending on the diversity of the sample.

3

Each interview is conducted individually, in a setting that ensures privacy and puts the participant at ease. The researcher follows the guide flexibly: asking the core questions, listening actively, and introducing probes ('Can you say more about that?', 'What did you mean by...?') to deepen responses. The conversation is audio- or video-recorded with the participant's consent. The researcher also takes brief field notes on non-verbal cues and contextual observations that the recording will not capture.

4

Recordings are transcribed verbatim, preserving the participant's own words, pauses, and relevant non-verbal markers (e.g., laughter, long silences). Transcription is checked against the original recording for accuracy. Each transcript is assigned a pseudonym or code to protect participant confidentiality. The resulting corpus of text becomes the dataset for analysis.

5

Transcripts are analysed using a method matched to the research question — thematic analysis, interpretive phenomenological analysis, grounded theory coding, narrative analysis, or content analysis. Regardless of the specific approach, the analyst reads transcripts repeatedly, assigns codes to meaningful segments, groups codes into higher-order themes or categories, and interprets patterns across participants. Negative cases (accounts that diverge from emerging patterns) are actively sought to refine and test interpretations.

6

Findings are written up with direct quotations from participants as evidential anchors, illustrating each theme or category. The researcher describes the analytic process transparently — how codes were developed, how themes were refined, and how rival interpretations were considered — so that readers can evaluate the rigour of the analysis. Transferability to other contexts is discussed rather than statistical generalisability claimed.

Common pitfalls

  • Leading questions — phrasing that implies an expected answer ('Don't you think that...?') contaminates participants' accounts and undermines the validity of the data; all questions must be genuinely open-ended.
  • Premature closure — stopping recruitment before saturation because the researcher feels they 'have enough data'; this often leads to findings that overlook important sub-groups or divergent perspectives.
  • Insufficient probing — accepting brief or vague responses without follow-up produces thin data that cannot sustain rigorous analysis; skilled probing ('Can you give me an example?', 'What happened next?') is essential.
  • Analytic drift — allowing interpretation to be driven by the most memorable or emotionally salient quotes rather than by systematic coding across the full dataset; every transcript must receive equal analytic attention.
  • Neglecting reflexivity — failing to document and account for the researcher's own positionality, expectations, and relationship to the topic can silently bias both data collection and interpretation.

Frequently asked questions

What is the difference between a structured, semi-structured, and unstructured interview?

A structured interview follows a fixed set of questions in a fixed order, identical for every participant — essentially an oral survey. A semi-structured interview uses an interview guide with topic areas and suggested questions, but the researcher adapts the order and wording in response to the conversation. An unstructured interview begins with one or two broad opening questions and follows entirely from the participant's responses. Most in-depth qualitative interviews are semi-structured: flexible enough to follow participants' accounts but focused enough to address the research question.

How many participants do I need?

There is no universally correct number. The guiding criterion is theoretical or informational saturation — the point at which new interviews are not introducing new themes or conceptual categories. In practice this commonly occurs between 10 and 30 participants, though highly homogeneous samples may saturate earlier and heterogeneous or maximum-variation samples may require more. Declaring a sample size in advance without a saturation rationale is not considered best practice in qualitative research.

Can I conduct the interviews online or by telephone?

Yes. Telephone and video-mediated interviews (via Zoom, Teams, etc.) are now widely accepted in qualitative research and can reduce geographic and logistic barriers. The trade-off is a partial loss of non-verbal cues available in face-to-face interviews. Researchers should note the mode of data collection in their methods section and consider whether it affected participant disclosure, particularly for sensitive topics.

How do I analyse in-depth interview data?

The choice of analytic method should follow from the research question and theoretical framework. Thematic analysis (Braun & Clarke) is the most widely used general approach and suits a broad range of questions. Interpretive phenomenological analysis is appropriate when the question concerns lived experience and meaning. Grounded theory coding is used when the goal is theory generation about a social process. Narrative analysis is appropriate when the structure and sequence of participants' stories are themselves the analytic focus. The in-depth interview is a data-collection method, not an analytic method, so it must be paired with a specified analytic approach.

Is an in-depth interview the same as a focus group?

No. A focus group brings multiple participants together simultaneously and uses group interaction — agreement, disagreement, storytelling, and co-construction of meaning — as part of the data. An in-depth interview is one-to-one and elicits the individual's perspective without peer influence. Focus groups are more efficient for surfacing shared social norms and collective views; in-depth interviews are preferable for sensitive topics, for capturing individual variation, or when the researcher wants to follow a single person's account in depth without group dynamics shaping responses.

Related methods

Sources

  1. Kvale, S. (1996). InterViews: An Introduction to Qualitative Research Interviewing. Sage.

    ISBN: 978-0803958203
    Copied.
  2. Rubin, H. J., & Rubin, I. S. (2005). Qualitative Interviewing: The Art of Hearing Data (2nd ed.). Sage.

    Open publication reference
    Copied.