How Should Your Planned Qualitative Analysis Inform Interview Questions?

How Do You Write Interview Questions That Will Support Your Planned Analysis?

You have selected a qualitative analysis approach and are preparing an interview guide. You know what your study aims to investigate, but how can you ensure that participants' responses will provide the evidence needed for analysis?

A common difficulty is developing interview questions directly from the research objectives without considering what the eventual analysis will require. The questions may be relevant to the topic but produce responses that are too general, superficial, or disconnected from the intended analytic purpose.

The opposite problem is equally consequential. Researchers may construct questions so closely around predetermined concepts that participants have little opportunity to introduce unexpected experiences or interpretations.

The challenge is designing interviews that generate rich, relevant evidence while preserving participants' capacity to describe their experiences in their own terms.

Design Interview Questions Around the Evidence Your Analysis Needs

Connecting Interview Design With Qualitative Analysis

Why Interview Questions Should Be Planned With Analysis in Mind

Qualitative interviews generate evidence through interaction between researchers and participants. The questions asked, the language used, the order of topics, and the interviewer's responses can influence the material produced.

Consequently, interview design is not independent of analysis. If your intended analysis requires detailed accounts of experiences, your questions must encourage participants to describe particular situations. If the analysis concerns processes, you need opportunities to explore actions, decisions, conditions, and changes over time.

For example, asking teachers whether they support generative AI may produce a general statement of approval or disapproval. Asking them to describe a recent occasion when they decided whether to use AI for a teaching task may reveal the considerations, constraints, and interpretations involved.

Neither question is inherently inappropriate. Their usefulness depends on what the study seeks to understand.

This relationship explains why selecting an initial qualitative analysis approach before data collection can improve interview design.

Distinguish Research Questions From Interview Questions

A research question identifies what the study seeks to understand. An interview question is a prompt used to generate material that may help answer the research question.

They should be related, but they do not need identical wording.

The research question contains an analytic concept, professional responsibility, that may not be part of participants' everyday vocabulary. The interview question invites a concrete account without requiring participants to adopt the researcher's terminology.

Follow-up questions can then explore who made decisions, how concerns were handled, and what responsibilities participants perceived.

Simply converting each research question into an interview question may produce abstract answers or encourage participants to reproduce the researcher's assumptions.

How Does the Analysis Approach Affect Interview Questions?

Different qualitative approaches emphasize different forms of evidence. Interview prompts should reflect those requirements.

These prompts are illustrative, not prescribed questions. Discourse-analytic traditions vary considerably, and some prioritize naturally occurring interactions rather than researcher-generated interviews.

Similarly, thematic analysis can be conducted through different theoretical orientations, while grounded theory and phenomenological approaches have methodological requirements extending beyond interview wording.

How Should Interview Questions Support Inductive Analysis?

When the analysis is predominantly inductive, interview questions should provide opportunities for participants to introduce experiences, concerns, and meanings that the researcher has not anticipated.

Broad opening questions can help establish participants' perspectives. However, openness does not mean avoiding specific follow-up questions.

For example, a researcher might begin by asking, "Can you tell me about your experiences using AI in teaching?"

When a participant mentions difficulties with assessment, the interviewer could ask, "Could you describe a particular situation when that difficulty arose?"

The follow-up encourages greater specificity without imposing a predetermined explanation.

Inductive analysis is not equivalent to conducting an interview without direction. The questions still need to generate material relevant to the research purpose.

How Should Interview Questions Support Deductive Analysis?

Deductive analysis may use concepts from existing theory or prior research to establish topics that should be explored during interviews.

Suppose a study examines teachers' adoption of educational technology using constructs concerning perceived usefulness and ease of use.

The interview guide may include prompts about how teachers evaluate the benefits of the technology and the effort required to use it.

However, researchers should avoid embedding theoretical conclusions in the questions.

For example, asking, "How has AI improved your teaching efficiency?" presupposes improvement. A more balanced question would be, "How, if at all, has AI affected the time or effort involved in your teaching tasks?"

The second formulation permits participants to describe positive, negative, mixed, or negligible effects.

The distinction between inductive and deductive coding is relevant because the source of analytic categories may influence which topics need to be addressed while leaving participants free to challenge those categories.

Should Every Planned Code Have a Corresponding Interview Question?

No. A direct one-to-one relationship between codes and interview questions is usually unnecessary and may be counterproductive.

A single interview response may contain material relevant to several codes. Conversely, a particular concept may become visible across multiple parts of an interview.

For example, a teacher describing an unsuccessful attempt to use AI in assessment may discuss reliability, accountability, workload, institutional guidance, and professional judgment within one account.

Requiring a separate question for every code can make interviews repetitive and overly structured.

For studies using a preliminary coding framework, it may be helpful to map broad interview topics to relevant conceptual domains. This is different from treating the interview as a checklist of codes that participants must confirm.

The decision to develop a codebook before data collection should not determine the entire conversational structure of the interview.

How Can You Elicit Richer Accounts Instead of Superficial Responses?

Analytically useful interviews often require more than participants' general opinions.

Researchers can invite participants to describe specific events, decisions, interactions, and consequences. Such accounts may provide greater contextual detail than broad evaluative statements.

Compare the following questions:

The alternatives encourage participants to provide examples and explanations. They do not guarantee rich responses, but they create opportunities for further probing.

Researchers should also recognize that participants may be unable or unwilling to provide detailed accounts. Interview quality depends partly on rapport, context, participant comfort, and the sensitivity of the topic.

What Role Should Follow-Up Probes Play?

Probes are supplementary questions or prompts that encourage clarification, elaboration, or reflection.

They are particularly useful when a participant introduces an analytically relevant issue but provides insufficient detail.

Common probing functions include:

• Clarification: "What did you mean when you described that decision as difficult?"
• Elaboration: "Could you tell me more about what happened?"
• Concrete example: "Can you recall a specific occasion when that occurred?"
• Sequence: "What happened after you made that decision?"
• Interpretation: "How did you understand that experience at the time?"
• Variation: "Has your experience been different in other situations?"

These prompts serve different analytic purposes. A narrative-oriented study may emphasize sequence and turning points, while an interpretive study may explore how participants make sense of particular experiences.

Probes should remain responsive to the participant's account rather than becoming a fixed series of questions applied mechanically.

How Should Interview Questions Be Sequenced?

Question sequence can influence how participants engage with the interview.

Beginning with accessible, experience-based questions may help participants become comfortable before discussing abstract or sensitive matters.

A semistructured interview might move from background experiences to specific incidents, followed by reflections on meanings, decisions, or changes over time.

However, the sequence should not be treated as universally fixed. Narrative interviews may benefit from allowing participants to establish their own chronology. Some phenomenological approaches may begin with a personally significant experience rather than general background questions.

The interviewer should also avoid introducing theoretical terminology too early when doing so might shape participants' subsequent descriptions.

Should Interview Questions Change During Data Collection?

They may, depending on the methodology and approved research procedures.

Early analysis can reveal that an important issue requires further exploration or that participants consistently misunderstand a question.

Researchers may refine wording, add probes, or reconsider the sequence of topics. In methodologies involving iterative data collection and analysis, these adjustments can be an integral part of the research process.

Grounded theory, for example, may involve progressively focused questioning as conceptual categories develop and further data are sought to elaborate them.

However, not every qualitative study requires iterative sampling informed by emerging analysis.

When interview questions change, researchers should document consequential revisions and consider whether earlier participants had comparable opportunities to discuss relevant topics.

Changes affecting approved participant procedures or sensitive topics may also require ethical review before implementation.

How Can You Avoid Leading Questions?

Leading questions suggest an expected answer or embed an assumption that participants may not share.

For example, "Why do teachers distrust AI-generated feedback?" presupposes distrust.

A less leading alternative is, "How do you evaluate AI-generated feedback when deciding whether to use it?"

Researchers should also avoid presenting only positive or only negative possibilities when the study aims to understand a range of experiences.

Nevertheless, no interview question is entirely neutral. Word choice, sequencing, interviewer identity, and the research context can influence responses.

The objective is not to eliminate every possible influence, but to construct questions thoughtfully and examine the researcher's role reflexively.

Should You Pilot Test the Interview Guide?

Pilot testing can help identify questions that are unclear, overly abstract, repetitive, or unlikely to produce sufficiently detailed accounts.

A preliminary interview may reveal whether participants understand the language used, whether the interview length is manageable, and whether important analytic topics are adequately explored.

Researchers can also examine a small amount of preliminary material using their intended analytic approach to assess whether the questions generate appropriate evidence.

However, a pilot interview cannot establish that the final analysis will produce particular themes or that all relevant experiences have been captured.

Pilot testing should be proportionate to the study and conducted within applicable ethical arrangements.

What Should Be Documented in the Interview Guide?

A practical interview guide may include the opening script, broad topic areas, principal questions, optional probes, transitions, and closing questions.

For research planning, an accompanying matrix may identify how each topic relates to the research question and intended analytic purpose.

This matrix is primarily a design aid. It should not force the interviewer to ask every question in an identical sequence unless the methodology requires such standardization.

Researchers may also maintain field notes and reflexive memos documenting how interview interactions, emerging interpretations, and methodological decisions influence the analysis.

Designing Interview Questions for a Study of Teachers' AI Decisions

Step 1: Identify the Evidence Required

The analysis needs detailed accounts of actual decisions, the circumstances surrounding them, and the meanings teachers attach to those experiences.

A simple question such as "Do you use AI for assessment?" would establish a practice but provide limited insight into the decision-making process.

Step 2: Develop Interview Topics and Prompts

Step 3: Preserve Opportunities for Unexpected Accounts

The researcher includes an open question near the end of the interview: "Is there anything about your experiences with AI-assisted assessment that we have not discussed but that you think is important?"

This provides an opportunity for participants to introduce issues outside the researcher's initial expectations.

Step 4: Review Early Interviews Analytically

After the first interviews, the researcher examines whether participants provide sufficiently detailed accounts of their decisions and interpretations.

Suppose participants frequently discuss informal conversations with colleagues. The researcher may add a follow-up prompt about how professional discussions influence assessment decisions.

This refinement is documented, and the researcher considers how differences between earlier and later interviews may affect the dataset.

Step 5: Develop Themes Through Analysis

The researcher does not automatically turn the interview topics into themes.

Instead, coding and interpretation may reveal patterns that cut across several topics. For example, participants might describe verification, institutional expectations, and personal judgment as interconnected aspects of professional responsibility.

Whether such a pattern becomes a theme depends on sustained analysis of the dataset, not on the number of interview questions devoted to it.

The practical advantage of this design is that the interview questions generate material relevant to the intended analysis while allowing participants to challenge the researcher's initial assumptions.

Common Mistakes When Designing Analysis-Aligned Interviews

How to Build an Interview Guide That Supports Your Analysis

Begin by identifying what the intended analysis must be able to examine. Then determine which experiences, events, meanings, or processes participants are positioned to describe.

Use these requirements to develop broad interview topics, followed by questions that invite concrete accounts and appropriate elaboration.

Once the interview guide is drafted, examine each principal question for relevance, clarity, neutrality, and potential analytic usefulness.

Ask whether the question invites participants to describe something they can reasonably know or recall. Consider whether the wording presupposes an outcome, combines several questions, or relies on technical terminology unfamiliar to participants.

Finally, remember that interviews are not the only source of qualitative evidence. If the intended analysis requires direct observations, naturally occurring interactions, or documentary records, reconsider whether interviewing alone can support the study's claims. The broader issue is aligning the analysis approach with appropriate data collection methods.

Before Finalizing Your Qualitative Interview Questions, Check:

Frequently Asked Questions About Planning Qualitative Interview Questions

Ask Questions That Generate Evidence, Not Predetermined Findings

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.