A structured interview asks every participant the same questions, in the same wording, in the same order, following a fixed administration protocol the interviewer is not free to depart from. It sits at the opposite end of the interview spectrum from the semi-structured interview, which is built around a flexible guide the interviewer can reorder and probe, and further still from the unstructured or narrative interview, which works from no predetermined question set at all. As CASRAI’s overview of qualitative data collection techniques notes, all three sit on a single continuum defined by how much latitude the interviewer has to depart from a script — structured interviews deliberately give up that latitude in exchange for comparability across participants, sites, and interviewers.
This guide covers when that trade is worth making, how to build a structured interview schedule, what standardisation actually protects against (interviewer effects, administration drift), and what it costs in interpretive depth. For the general landscape of interview types and when each suits a research question, see CASRAI’s overview of interviews as a research method; for building the flexible alternative, see the guide to designing semi-structured interviews.
What Makes an Interview “Structured”
A structured interview is defined by four features, all fixed in advance and applied identically to every participant:
- Fixed wording. Questions are asked verbatim from a written schedule — the interviewer cannot rephrase, simplify, or explain a question beyond a scripted clarification, because doing so for one participant and not another introduces variation the design is meant to eliminate.
- Fixed order. Questions are asked in the same sequence for every participant, because question order itself can shift how later questions are interpreted (an order effect), and an inconsistent order confounds comparison across respondents.
- Constrained or coded response formats. Many structured interviews use closed or semi-closed response categories precisely so responses can be tallied and compared the way survey responses are; where open-ended questions are retained, responses are typically coded against a predetermined scheme rather than analyzed thematically from scratch.
- A written administration protocol. Beyond the questions themselves, the interviewer follows fixed rules for how to open the interview, how to handle a “don’t know” or off-topic response, how much (if any) unscripted probing is permitted, and how responses are recorded.
Structured interviewing is, functionally, an interviewer-administered questionnaire. The presence of an interviewer changes the delivery mode and what it can accomplish (see the comparison to survey administration below), but it does not loosen the standardisation discipline — if anything, a live interviewer requires a stricter written protocol than a self-administered form does, because there is a person in the room who could otherwise introduce variation.
Structured vs. Semi-Structured vs. Unstructured Interviews
| Dimension | Structured | Semi-Structured | Unstructured |
|---|---|---|---|
| Question wording | Fixed, read verbatim | Guide-based, can be rephrased | No predetermined wording |
| Question order | Fixed for all participants | Can be reordered mid-interview | Emerges from the conversation |
| Probing / follow-up | Minimal or scripted only | Planned probes plus improvised follow-up | Fully improvised, participant-led |
| Response format | Often closed or precoded | Open-ended, coded after the fact | Open-ended, narrative |
| Interviewer training burden | Lower — administer the script correctly | Higher — judge when and how to probe | Highest — build rapport and follow emergent threads |
| Comparability across participants | High | Moderate | Low |
| Interpretive depth per participant | Low | Moderate to high | High |
| Typical analysis | Quantitative tallying or simple coding | Thematic analysis, framework analysis | Narrative or phenomenological analysis |
| Typical use case | Multi-site studies, large-N interview surveys, compliance/audit interviews | Most qualitative research interviews | Life-history, ethnographic, narrative inquiry |
When Standardisation Is the Right Call
Structure is a design decision, not a default, and it should be chosen for a specific reason rather than out of caution. It is usually the right call when:
- Multiple interviewers are collecting data. Any time more than one person is asking the questions, wording and order drift introduces a source of variance that is indistinguishable, in the resulting data, from a genuine difference between participants. Standardisation is the primary control against this.
- The study spans multiple sites. Cross-site comparability is only interpretable if “the same question” actually was the same question at every site — a translated or locally adapted semi-structured guide can drift in ways a fixed schedule is built to prevent.
- Responses will be analyzed quantitatively. If the plan is to tally, cross-tabulate, or run inferential statistics on interview responses, the responses need to be comparable across cases in the way survey data is, which requires fixed wording and, usually, coded response categories.
- The interview needs to be defensible after the fact. Compliance, audit, and some regulatory-inspection interviews (and structured witness or investigative interviews in a research-integrity inquiry) benefit from a documented, repeatable protocol that shows every respondent was asked the same thing in the same way — a semi-structured approach is harder to defend as procedurally fair if challenged.
- The study is a replication or a repeated wave. Longitudinal or trend studies that compare responses across time points need the instrument held constant across waves; a semi-structured guide that evolves naturally between rounds undermines that comparison.
Standardisation is generally the wrong call when the research question itself is about how participants frame or make sense of something in their own terms — that is the territory of semi-structured and unstructured interviewing, not structured.
Constructing a Structured Interview Schedule
- Derive every item directly from the research question or conceptual framework. Because there is no room to improvise once fieldwork starts, an item that turns out to be ambiguous or off-target cannot be fixed mid-study the way a semi-structured guide can absorb that correction on the fly.
- Fix the exact wording of every question. Write the question as it will actually be spoken, not as a topic prompt to be phrased in the moment — the schedule is the instrument, not a reminder of what to ask about.
- Fix the order and any branching logic. If certain questions only apply to some participants (skip logic), the branching rule itself needs to be written into the protocol and applied identically, not judged case by case.
- Decide the response format up front. Closed categories need to be exhaustive and mutually exclusive; if open-ended responses are retained, decide the coding scheme before fieldwork, not after, so every interviewer records responses in a form that can be coded consistently later.
- Write the administration protocol, not just the questions. Specify the scripted opening (purpose, consent, expected length), exactly what an interviewer may say if asked to clarify a question, how to handle silence or a “don’t know,” and how responses get recorded in the moment.
- Pilot with more than one interviewer, not just one. A single-interviewer pilot only tests whether the questions work; it cannot detect administration drift between interviewers, which is the specific risk a structured design exists to control. Piloting with the actual interview team surfaces places where the script is ambiguous enough that two trained interviewers would deliver it differently.
- Train and calibrate the interview team. Calibration — having interviewers practice against the same mock scenarios and comparing how each one administers and records them — is what actually produces standardisation in practice; a written protocol alone does not guarantee two interviewers will apply it the same way.
Inter-Interviewer Reliability: The Problem Standardisation Exists to Solve
Whenever more than one person collects the data, the study has to account for inter-interviewer (sometimes called inter-rater, in the coding stage) reliability — the degree to which different interviewers, administering the same instrument to comparable participants, produce comparable data. Left unmanaged, interviewer effects are a real and well-documented source of measurement error: differences in tone, pacing, willingness to clarify, and nonverbal cues can all shift how a participant answers, independent of anything true about the participant. See CASRAI’s comparison of test-retest vs. inter-rater reliability for how this concept relates to the broader reliability framework a study reports against.
A structured schedule and a written administration protocol are the primary controls, but they are a design choice, not a guarantee, and most structured-interview studies build in at least one additional check:
- Interviewer calibration before fieldwork (see above), repeated periodically on longer studies to catch drift.
- Audio recording a sample of interviews per interviewer so a supervisor can audit administration fidelity against the written protocol, not just the content of the answers.
- Double-coding a subsample of responses independently, when coded response categories are used, and reporting agreement between coders.
- Comparing response distributions across interviewers as a diagnostic — a systematic difference in how one interviewer’s respondents answer, holding sampling constant, is a flag for administration drift rather than a genuine finding.
What Standardisation Costs You
The trade a structured interview makes is real, not just a design footnote. Fixed wording and fixed order mean the interviewer cannot follow an unanticipated but important thread the way a semi-structured interview is built to allow — a participant who introduces a theme the schedule never anticipated gets it noted, at best, as an aside, not explored. Coded or closed response formats can force a participant’s actual answer into the nearest available category rather than capturing it in their own terms, which is precisely the loss of nuance that qualitative interviewing is usually chosen to avoid. And a rigidly scripted interaction can read to participants as impersonal, which affects rapport and, in sensitive-topic research, can affect disclosure.
None of this makes structure the wrong choice where comparability genuinely matters — it means structure should be chosen deliberately, for a reason tied to the analysis plan, and not as a default because a fixed script feels more rigorous. Where the research question is really about how participants make sense of something in their own frame of reference, that loss of depth is usually too costly, and a semi-structured design (see CASRAI’s guide to designing semi-structured interviews) is the better fit.
Structured Interviews vs. Survey Administration
A structured interview and an interviewer-administered survey are, in most methodological respects, the same instrument delivered the same way — the boundary between the two terms is more disciplinary convention than methodological substance. What genuinely distinguishes a structured interview from a self-administered questionnaire (see CASRAI’s guide to questionnaire design and its overview of survey research methods) is the presence of a live interviewer:
- Lower item non-response. An interviewer can note and gently follow up on a skipped question in a way a paper or online form cannot.
- Controlled clarification. A scripted, permitted clarification can resolve a confusing question without abandoning standardisation, something a self-administered form has no mechanism for.
- Higher cost per response. Interviewer time is the main reason structured interviews are typically fielded at far smaller scale than self-administered surveys, which is itself a reason to reserve the format for research questions where interviewer presence adds real value.
- Susceptibility to social-desirability and interviewer-presence effects. A live interviewer can improve completion and clarity but can also shift how a participant answers a sensitive question compared with an anonymous self-administered form — a trade-off worth naming explicitly in a schedule’s design, particularly for sensitive-topic research.
Where researchers are choosing question wording rather than interview structure itself, CASRAI’s guide to survey question types covers the same closed-versus-open-ended trade-offs that apply to a structured interview schedule.
Frequently Asked Questions
What is a structured interview in research?
A structured interview is a data-collection method in which every participant is asked the same set of questions, in the same wording and order, following a fixed administration protocol — functionally an interviewer-administered questionnaire, chosen when comparability across participants, interviewers, or sites matters more than the flexibility to probe unanticipated themes.
How is a structured interview different from a questionnaire?
The instrument and the standardisation discipline are essentially the same; the difference is delivery. A structured interview is delivered by a live interviewer following a script, which can reduce item non-response and allow scripted clarification, while a questionnaire is typically self-administered. See the comparison to survey administration above.
Can structured interview data be analyzed qualitatively?
To a limited degree. If open-ended questions are retained, responses can be coded and even thematically summarized, but the fixed wording and order — and the lack of participant-led probing — generally produce thinner, less contextually rich data than a semi-structured or unstructured interview, so structured-interview data is more often analyzed by tallying or simple coding than by full thematic or narrative analysis.
How many interviewers does it take before standardisation matters?
As soon as more than one interviewer is collecting data, interviewer effects become a real source of variance that a fixed schedule and administration protocol are designed to control — this is true even with just two interviewers, and the risk grows with team size and with the length of the fieldwork period, which is why calibration is usually repeated periodically rather than done once at the start.
Should I use a structured or semi-structured interview?
Choose structured when comparability across participants, interviewers, or sites is central to the analysis plan — multi-site studies, large-N interview surveys destined for quantitative analysis, or compliance and audit interviews. Choose semi-structured, covered in CASRAI’s guide to designing semi-structured interviews, when the research question is about how participants make sense of something in their own terms and interpretive depth matters more than strict comparability.
