Conducting In-Depth Interviews

Overview

An in-depth qualitative interview is a purposeful conversation designed to explore how a participant understands, experiences or explains a phenomenon. It differs from a structured survey interview because the participant is not merely a source of standardised responses. Meaning is developed through interaction, clarification, examples and follow-up questions.[1]

Semi-structured interviewing is common in health research because it balances consistency across participants with enough flexibility to pursue unexpected but relevant experiences. Rigour depends on more than asking the same questions: researchers need an aligned interview guide, trained interviewing, reflexivity, careful recording and an ethical plan for sensitive material.[2,3]

Why this matters

Poor interviewing can generate superficial data that no analytic technique can repair. Leading questions may reproduce the researcher’s assumptions; rapid questioning can prevent reflection; clinical interviewing habits may turn research into advice-giving; and participants may withhold criticism when the interviewer is also their clinician.

Conversely, a skilful interview can reveal sequence, context, contradictions and meaning that structured response options cannot capture.[1,3]

When to use in-depth interviews

DECISION GUIDE

Choose individual interviews when

• The study seeks detailed personal experience or individual trajectories.
• Topics are sensitive, private or difficult to discuss in a group.
• Participants may have unequal status or conflict that would constrain group discussion.
• The researcher needs to follow each person’s chronology or reasoning in depth.
• Participants are geographically dispersed and remote interviewing is feasible.
• The question does not depend primarily on observing real-time social interaction.

Interviews are less appropriate when the primary interest is what people actually do in a setting rather than how they describe it; observation may then be necessary. Focus groups may be preferable when interaction among participants is itself informative.

Developing the interview guide

Kallio and colleagues proposed a staged approach to semi-structured guide development that includes establishing whether semi-structured interviews fit the purpose, using prior knowledge, formulating the preliminary guide, pilot testing and presenting the final guide.[2]

A guide should be organised around the research question rather than around a questionnaire-style list of variables.

PROTOCOL CHECKLIST

Build a coherent semi-structured interview guide

• Begin with one or two broad, accessible opening questions.
• Organise prompts around conceptual domains linked to the research aim.
• Prefer open questions that invite description and examples.
• Avoid double-barrelled, leading and judgemental wording.
• Prepare neutral probes for clarification, sequence, meaning and concrete examples.
• Include transitions into sensitive topics rather than introducing them abruptly.
• End with an opportunity for participants to add issues the guide missed.
• Pilot the guide for comprehension, flow, burden and analytic usefulness.

Illustrative interview-guide sequence

illustrative

Table 1. From broad experience to focused probes

Purpose	Main question	Possible probes
Open the account	Tell me about the period after you left maternity care.	What changed? What became most important day to day?
Explore risk meaning	What did future diabetes risk mean to you after the pregnancy ended?	Did that meaning change over time?
Explore pathway	Can you take me through what happened with the recommended glucose test?	Who contacted you? What happened next?
Explore constraints	What made follow-up easier or harder?	Travel, childcare, work, cost, clinic arrangements?
Explore improvement	If you could redesign the follow-up process, what would you change?	Which change would matter most?

The probes shown in a guide are prompts, not a checklist that must be read mechanically to every participant.

Conducting the interview

Establish the interview context

The researcher should explain the study, confirm consent, discuss recording, clarify confidentiality limits and establish how long the interview is expected to take. The location—physical or virtual—should provide privacy and minimise interruptions.

Build rapport without manufacturing agreement

Rapport helps participants speak openly, but the interview is not a friendship or therapeutic encounter. The interviewer should be warm, attentive and respectful while avoiding exaggerated agreement that signals preferred answers.[3]

Ask one question at a time

Broad questions invite narrative. Follow-up questions then deepen the account:

• Clarification: “When you say the clinic was difficult, what made it difficult?”
• Example: “Can you describe a particular time when that happened?”
• Sequence: “What happened after you received the message?”
• Meaning: “What did that suggest to you about your risk?”
• Contrast: “Was that different from your experience during pregnancy?”

Silence can be productive. Filling every pause may interrupt reflection.

Follow the participant without losing the research question

Semi-structured interviewing is responsive, not aimless. The interviewer may pursue unexpected material when it helps answer the study question, while gently redirecting prolonged tangents.

Interviewer influence and reflexivity

Interviewer characteristics can influence disclosure and interpretation. A clinician may understand medical shorthand but may unintentionally trigger deference. A researcher who experienced the same condition may build rapport but assume shared meaning too quickly.

McGrath and colleagues emphasise preparation, reflexivity, relationship building, suitable questioning and attention to the interview environment.[3]

Reflexive notes after interviews can record:

• moments when the interviewer led or interrupted;
• assumptions that were challenged;
• emotional responses;
• changes in the guide and why;
• apparent effects of professional or social position; and
• emerging ideas to examine in later interviews.

Recording, transcription and field notes

Audio recording supports detailed analysis but requires consent and secure storage. Researchers should specify whether transcription is verbatim, whether non-verbal features are included, who transcribes, how recordings are checked and how translation is handled.

Field notes made immediately after the interview can capture context not obvious in the recording: interruptions, setting, rapport, notable emotional shifts and the researcher’s initial reflections. Field notes should be distinguished from the participant transcript.

Remote interviews

Telephone or video interviews can improve geographic access and may offer privacy or convenience, but they change the interaction. Video requires bandwidth and a private location; telephone interviews remove visual cues; online platforms introduce data-security considerations. The protocol should explain why the selected mode fits the participants and topic rather than assuming remote interviewing is methodologically identical to face-to-face interaction.

Sensitive topics and distress

Qualitative interviews may surface grief, trauma, stigma, domestic conflict, professional misconduct or fear about illness. Dempsey and colleagues emphasise preparation for sensitive interviews, including interviewer competence, gatekeepers, setting, participant vulnerability and appropriate disengagement.[4]

Pilot interviewing and interviewer training

Pilot interviews test whether questions elicit material capable of answering the study question. A guide can be grammatically clear yet analytically poor. Pilot work should examine:

• whether participants understand the questions;
• whether questions invite examples rather than short opinions;
• whether domains appear in a logical order;
• whether sensitive transitions are acceptable;
• interview duration;
• interviewer use of probes; and
• whether generated data have sufficient depth.

Interviewers should practise with feedback. In multicentre or multi-interviewer studies, standardisation concerns the study purpose and ethical conduct, not forcing every conversation into identical wording.

Analysis interface

Interviewing and analysis should be connected. Early analysis can identify areas needing deeper probing, contrasting participants or clarification. This is especially important in grounded theory and other iterative designs. Researchers should document substantive guide changes rather than presenting the final guide as if it existed unchanged from the start.

Interview interaction is part of the data-generating contextMMed extension

From an interpretive perspective, an interview does not merely extract an account that already existed in fixed form. The participant responds to the particular questions, interviewer and setting. This does not make the data invalid; it means the research relationship should be considered during analysis and reporting.[1]

Worked example

WORKED EXAMPLE

Revising a judgemental postpartum interview

The initial question, “Why did you fail to attend the glucose test even though you knew it was important?” contains three assumptions: that attendance is solely the woman’s responsibility, that she knew the test was important, and that non-attendance represents failure.

The revised sequence begins: “Tell me what happened with your health care after the baby was born.” The interviewer then asks, “What, if anything, were you told about diabetes risk after pregnancy?” and “Can you take me through what happened with the recommended glucose test?” If the participant raises transport, childcare or fear, the interviewer explores those issues with examples rather than introducing them prematurely.

After the first interviews, the team notices that several women distinguish maternity care from “my own health care”. They add a neutral probe about this transition and document the change in reflexive memos. The modification is part of iterative qualitative data generation rather than a concealed protocol failure.

Suggested protocol or reporting wording

Interview mode, language and setting

Interview mode changes the research interaction. Face-to-face interviews provide access to embodied and contextual cues but may require travel and may feel less private in a clinical setting. Telephone interviews can reduce travel and visual social pressure, while video interviews preserve some visual interaction but depend on technology and private space. The choice should be justified by the population, sensitivity of the topic, access requirements and research question rather than by convenience alone.[1,3]

The setting also communicates power. Interviewing a patient in the clinician’s consultation room may reinforce a clinical hierarchy even when the researcher says the interview is independent. A neutral private room or participant-selected secure setting may support more open discussion.

Multilingual interviewing

When interviews occur in several languages, translation becomes part of methodology. The protocol should state who conducts each language interview, whether transcripts are produced in the original language, when translation occurs, and how culturally specific meanings are handled. Translating every transcript before analysis can improve team accessibility but may lose nuance; analysing in the original language can preserve meaning but requires analysts competent in that language.

A practical strategy is to maintain key terms in the original language within analytic memos when no direct English equivalent exists, and to discuss translation choices when selecting quotations for publication.

Training and quality assurance for interview teams

A semi-structured guide cannot compensate for an interviewer who reads questions mechanically or fails to probe. Team preparation should include role-play, discussion of the research purpose, practice with neutral probes, handling silence, responding to distress and reflection on interviewer assumptions.[2-4]

Early recordings can be reviewed for training—not to make every interviewer sound identical, but to identify leading questions, premature topic changes, excessive researcher talk or missed opportunities for elaboration. The team should document major changes to the guide or interviewer practice.

PROTOCOL CHECKLIST

Interview-team quality assurance

• Train interviewers in the study question and qualitative orientation, not only the question list.
• Practise open prompts, follow-up probes and tolerating silence.
• Discuss power relationships and social-desirability risks.
• Review early interviews and field notes for depth and leading language.
• Hold periodic debriefings as data collection and analysis evolve.
• Document substantive guide revisions and why they were made.
• Plan consistent procedures for recording, naming, transfer and secure storage of audio files.
• Define how translated interviews and quotations will be checked.

Knowing when an interview is sufficiently rich

Duration is not a proxy for quality. A 90-minute interview can remain superficial, while a focused 35-minute interview may generate rich material. Researchers should attend to specificity, examples, reflection, contradictions and contextual detail. Questions such as “Can you take me through what happened that day?”, “What made that difficult?”, “How did you decide what to do next?” and “Was it always like that?” often move an account from opinion to experience.[3]

After each interview, a short debrief can record what was learned, what remained unclear, how the interviewer may have influenced the discussion and what should be explored in later interviews. This makes data generation genuinely iterative rather than a production line of identical encounters.

An interview is a research relationship, not a questionnaire delivered orallyMMed extension

Semi-structured interviewing permits consistency of purpose without requiring identical wording or sequence. The interviewer follows meaning, asks for examples and notices tensions while still ensuring that core areas relevant to the question are covered. This is why interview quality depends on judgement, reflexivity and listening as much as on guide construction.

Closing the interview well

The final minutes of an interview are methodologically and ethically important. Invite participants to add anything the guide missed, clarify what will happen to the recording, and provide an appropriate route for questions or support. Researchers should also document any relevant post-interview context in field notes before memory fades.

Reporting interview procedures

For interview-based studies, COREQ highlights reporting of the research team and reflexivity, study design, and analysis and findings.[5] Use it to check whether readers can understand who conducted interviews, participant relationships, setting, recording, field notes and analytic procedures; do not treat checklist completion as proof that the interviews were methodologically strong.

Common mistakes

Knowledge check

Why is 'Did transport difficulties prevent you from attending?' a weak opening question?

Answer and reasoning

It introduces a specific explanation before learning whether transport mattered to the participant. A broader question about what happened with follow-up allows transport or other barriers to emerge, after which targeted probes can clarify their importance.

What is the purpose of probing in a qualitative interview?

Answer and reasoning

Probing deepens understanding by asking for clarification, examples, sequence, meaning or contrast. It should extend the participant’s account rather than steer the participant towards the researcher’s preferred explanation.

Why can silence be useful?

Answer and reasoning

Participants may need time to recall events, decide how to express something or reflect on sensitive material. Immediate follow-up by the interviewer can interrupt that process and shorten the account.

A participant becomes upset while discussing pregnancy loss. What should the interviewer do?

Answer and reasoning

Follow the study’s distress protocol: acknowledge the distress, offer to pause or stop, check the participant’s wishes, and use the predefined support or referral pathway when indicated. The interviewer should not improvise promises or clinical treatment outside the approved role.

Key take-home messages

SUMMARY

Key take-home messages

• Individual interviews are best when detailed personal accounts and privacy are central.
• Build the guide around the research question, then use open questions and neutral probes.
• Interview quality depends on listening, responsiveness and reflexivity, not rigidly identical wording.
• Plan recording, transcription, field notes, sensitive disclosures and remote interviewing prospectively.
• Connect data generation with iterative analysis and report meaningful guide changes transparently.

References

1. DiCicco-Bloom B, Crabtree BF. The qualitative research interview. Med Educ. 2006;40(4):314-321. https://doi.org/10.1111/j.1365-2929.2006.02418.x
2. Kallio H, Pietilä AM, Johnson M, Kangasniemi M. Systematic methodological review: developing a framework for a qualitative semi-structured interview guide. J Adv Nurs. 2016;72(12):2954-2965. https://doi.org/10.1111/jan.13031
3. McGrath C, Palmgren PJ, Liljedahl M. Twelve tips for conducting qualitative research interviews. Med Teach. 2019;41(9):1002-1006. https://doi.org/10.1080/0142159X.2018.1497149
4. Dempsey L, Dowling M, Larkin P, Murphy K. Sensitive interviewing in qualitative research. Res Nurs Health. 2016;39(6):480-490. https://doi.org/10.1002/nur.21743
5. Tong A, Sainsbury P, Craig J. Consolidated criteria for reporting qualitative research (COREQ): a 32-item checklist for interviews and focus groups. Int J Qual Health Care. 2007;19(6):349-357. https://doi.org/10.1093/intqhc/mzm042