03 · What You Need to Know
How to Determine Whether Qualitative Research Fits Your Question
What Makes Research Qualitative?
Qualitative research investigates phenomena through evidence that allows researchers to examine meaning, experience, social practices, relationships, or processes in context.
Its materials may include interviews, observations, documents, field notes, diaries, recorded interactions, images, and other forms of evidence appropriate to the question.
What distinguishes qualitative inquiry is not simply the use of words instead of numbers. It is the relationship between the research question, the evidence, and the interpretive or explanatory work undertaken during analysis.
For example, asking students to explain how they decide whether to accept AI-generated feedback may produce evidence about their reasoning and practices. A qualitative analysis could investigate patterns across those accounts while preserving important differences between participants.
Qualitative research is not one uniform methodology. Different traditions make different assumptions about knowledge, experience, social reality, and the researcher's role in interpretation.
Consequently, choosing a qualitative approach is only the beginning of the methodological decision.
Which Research Questions Are Particularly Suitable for Qualitative Inquiry?
Qualitative research is especially useful when the intended answer requires examining how participants understand a phenomenon, how practices unfold, or how processes operate within particular circumstances.
| Research purpose |
Illustrative question |
Potential qualitative contribution |
| Understand experiences |
How do doctoral students experience uncertainty during dissertation writing? |
Examines the meanings and circumstances associated with participants' experiences. |
| Investigate interpretations |
How do teachers interpret institutional expectations concerning generative AI? |
Identifies how participants understand and negotiate particular expectations. |
| Examine practices |
How do students evaluate AI-generated explanations while completing assignments? |
Investigates actual or reported practices and the reasoning associated with them. |
| Investigate processes |
How do academic departments introduce AI-supported assessment practices? |
Examines decisions, interactions, and developments over time. |
| Develop conceptual explanations |
How do researchers develop changing conceptions of scholarly success? |
Supports systematic development of concepts or theoretical explanations. |
| Understand organizational culture |
How are expectations about technology use established within a faculty community? |
Examines shared practices, relationships, and meanings in their social setting. |
These questions share an interest in understanding phenomena rather than merely estimating their numerical distribution.
However, their differences matter. Investigating an individual's lived experience is not methodologically identical to studying organizational culture or developing a theory of social processes.
The particular qualitative tradition should therefore follow the intended contribution.
When Does a Question Require Experiential Evidence?
Experiential evidence becomes important when the phenomenon concerns how people perceive, interpret, remember, or make sense of particular circumstances.
For example, a researcher might investigate how university instructors experience the introduction of mandatory AI-assisted assessment procedures.
A survey could estimate the proportion of instructors reporting positive or negative attitudes. Interviews might reveal how instructors understand the policy, negotiate professional responsibilities, and interpret changes to their teaching practices.
These are different forms of evidence.
Qualitative accounts can illuminate experiences that predefined response categories may not anticipate. They can also reveal tensions or ambiguities that are obscured when participants must select a single response option.
Nevertheless, participants' accounts should not automatically be treated as complete records of everything that happened. Memory, social expectations, interview interactions, and the researcher's interpretation can influence the evidence.
When the central question concerns evidence about individual experiences, the researcher should consider how those experiences can be investigated with sufficient depth and methodological coherence.
When Is Qualitative Research Appropriate for Studying Processes?
Qualitative approaches can be especially useful when the research question concerns how a sequence of decisions, interactions, or activities unfolds.
Suppose a university introduces a policy governing the use of generative AI in assessment.
The researcher may want to understand how departments interpret the policy, how instructors modify assessment tasks, and how students respond to the resulting requirements.
Interviews, institutional documents, observations, and records of implementation decisions may help reconstruct these developments.
Depending on the design, qualitative evidence may support an account of how the policy was enacted and which processes shaped its implementation.
However, reconstructing a process does not automatically establish that a particular event caused the final outcome. Causal explanations require systematic consideration of temporal ordering, competing explanations, and the inferential strength of the evidence.
The distinction is especially important when moving from describing a process to claiming that the process generated an outcome.
Can Qualitative Research Explain Why Something Happened?
Yes. Qualitative research can contribute to explanations of actions, decisions, organizational developments, and causal processes.
For example, a case study might investigate why an educational technology initiative was adopted in one department but resisted in another.
The researcher could examine institutional documents, interview decision-makers, and reconstruct relevant events to assess plausible explanations.
Some qualitative approaches explicitly investigate causal mechanisms, while others focus primarily on meaning, interpretation, or conceptual understanding.
These purposes should not be conflated.
Participants' reported reasons for an action may provide important evidence about their intentions. They do not necessarily establish every causal influence on that action.
When the research question asks why something happened rather than merely whether it happened, the researcher must clarify what form of explanation is required.
Which Qualitative Approach Should You Consider?
Qualitative traditions differ in their objectives, theoretical assumptions, and analytical procedures. The following comparison identifies broad orientations rather than exhaustive definitions.
| Approach |
Typical focus |
Illustrative question |
| Phenomenological inquiry |
Experiences and the meanings associated with a phenomenon |
How do doctoral students experience academic uncertainty? |
| Grounded theory |
Developing an explanatory theoretical account grounded in systematic engagement with data |
How do researchers negotiate changing definitions of scholarly success? |
| Ethnography |
Social practices, cultural meanings, and relationships within a group or setting |
How do faculty members establish informal norms concerning AI use? |
| Case study |
In-depth investigation of a bounded case or cases within their circumstances |
How did one university implement an institution-wide AI assessment policy? |
| Narrative inquiry |
Experiences and identities as understood or communicated through stories |
How do teachers narrate their professional transitions into AI-supported instruction? |
| Qualitative descriptive inquiry |
Providing a systematic account of experiences, events, or practices close to participants' descriptions |
What challenges do instructors report when introducing AI-supported feedback? |
These approaches are not interchangeable labels for interview-based research.
For example, conducting interviews does not automatically make a study phenomenological. A grounded theory study generally requires more than grouping interview responses into themes. Ethnography ordinarily involves sustained attention to social practices and cultural context rather than merely surveying members of a group.
Researchers should examine the methodological literature associated with the selected tradition and explain how their procedures reflect its purpose.
Does Qualitative Research Always Require Interviews?
No. Interviews are common, but they are not a defining requirement.
Observation may be particularly useful when the question concerns practices that participants find difficult to describe or may not recognize explicitly.
Document analysis can investigate institutional policies, historical decisions, or the ways organizations formally represent particular issues.
Diaries may provide evidence about experiences as they unfold over time. Recorded interactions can reveal conversational practices or collaborative decision-making.
Digital traces, such as successive document revisions or discussion records, may also contribute to qualitative analysis when interpreted in relation to an appropriate research question.
For example, studying how students negotiate the use of AI-generated feedback might involve observing writing sessions and examining successive drafts, rather than relying exclusively on retrospective interviews.
The evidence source should be selected according to what the researcher needs to observe or understand.
How Should Participants or Cases Be Selected?
Qualitative sampling is commonly guided by the relevance of participants or cases to the research question rather than by a requirement for statistical representativeness.
Purposive sampling may identify participants with particular experiences, roles, or characteristics.
Maximum variation sampling may help examine differences across relevant circumstances. Criterion-based sampling may focus on participants who meet clearly defined conditions.
Some grounded theory traditions use theoretical sampling, in which emerging analysis informs subsequent decisions about what evidence or cases to seek.
Convenience sampling is not automatically invalid, but researchers should consider whether it provides access to the experiences or processes necessary for the intended contribution.
For example, interviewing only enthusiastic early adopters may provide limited evidence for a question about why some teachers resist educational technology.
The selection strategy should therefore reflect the conceptual or explanatory needs of the study.
How Many Participants Does Qualitative Research Need?
There is no universal participant count that makes a qualitative study adequate.
Sample adequacy depends on the question's scope, participant specificity, methodological orientation, quality of evidence, and intended analysis.
Malterud and colleagues proposed the concept of information power, suggesting that the more relevant information a sample provides for a particular study, the fewer participants may be needed. This is a planning consideration rather than a universal numerical formula.
Saturation is also used in some qualitative traditions, but its meaning and suitability vary. Researchers should specify what kind of saturation they mean rather than treating the term as an automatic justification for stopping recruitment.
Some interpretive approaches do not regard saturation as an appropriate universal criterion, particularly where the analysis treats meaning as contextually produced rather than as a finite set of discoverable categories.
A small, carefully selected sample may be appropriate for an intensive interpretive study. A broader investigation involving several institutions and participant groups may require substantially more evidence.
The relevant question is whether the sample and data provide an adequate basis for the intended analytical contribution.
What Makes Qualitative Analysis Rigorous?
Qualitative rigor depends on methodological coherence, quality of evidence, analytical transparency, and the credibility of the relationship between observations and interpretations.
Different traditions emphasize different quality criteria. There is no single checklist that can establish rigor across all qualitative approaches.
Nevertheless, several considerations are widely relevant.
Researchers should explain how data were generated, how participants or cases were selected, how analytical decisions were made, and how interpretations relate to the evidence.
They should also consider contradictory observations, alternative interpretations, and the influence of their own assumptions and positions.
Reflexivity involves examining how the researcher's background, relationships, commitments, and decisions may shape the research process. It is not merely a declaration that bias exists.
Documentation of analytical decisions can strengthen transparency, but an extensive audit trail does not compensate for superficial analysis.
Likewise, using qualitative analysis software does not automatically make coding or interpretation rigorous.
Is Thematic Analysis Appropriate for Every Qualitative Study?
No. Thematic analysis is a family of approaches concerned with identifying and interpreting patterns of meaning across qualitative evidence.
It can be suitable for many research questions, but its assumptions and procedures should be consistent with the study's methodological orientation.
For example, reflexive thematic analysis, as developed by Braun and Clarke, emphasizes the active interpretive role of the researcher. Coding reliability approaches may instead emphasize consistency between coders under a different understanding of coding and theme development.
These approaches should not be combined indiscriminately.
A study using reflexive thematic analysis should not automatically treat inter-rater agreement as a necessary indicator of quality. Conversely, a study designed around standardized coding may have legitimate reasons to examine coding consistency.
The researcher should specify the version of thematic analysis being used and explain why it fits the question and analytical purpose.
Can Qualitative Research Produce Generalizable Knowledge?
Qualitative research can produce findings that inform understanding beyond the immediate participants or setting, but the nature of that contribution differs from statistical generalization.
Statistical generalization typically concerns inference from a sample to a defined population under appropriate sampling assumptions.
Qualitative studies may instead support theoretical development, analytical generalization, or judgments about transferability, depending on their methodological orientation.
For example, a detailed study of how instructors negotiate institutional AI policies may identify concepts or processes relevant to other educational organizations.
Whether those insights apply elsewhere depends on the theoretical argument, evidence, and relevant similarities or differences between settings.
Rich contextual description can help readers assess transferability, but it does not automatically establish that the findings represent the prevalence of experiences in a broader population.
Researchers should describe the scope of their conclusions precisely rather than claiming either universal applicability or complete context-specificity without justification.
Can Qualitative Research Establish Causal Explanations?
Some qualitative approaches can contribute substantially to causal explanation.
Process tracing, for example, may investigate whether sequences of events and within-case observations support particular causal mechanisms over competing explanations.
Qualitative comparative inquiry can also examine how combinations of conditions relate to outcomes, although the strength of causal conclusions depends on case selection, analytical procedures, and assumptions.
However, not every qualitative study is designed for causal inference.
Interview accounts describing why participants believe something happened may be informative without establishing that those beliefs identify the actual causal process.
Researchers should distinguish explanatory plausibility from causal identification and specify the evidence needed for the intended claim.
When the central question requires evidence about the mechanisms producing an outcome, qualitative inquiry may be appropriate if its design can investigate those mechanisms systematically.
When Might Qualitative Research Be a Poor Fit?
Qualitative inquiry may be insufficient when the central research question requires numerical estimates that cannot be established through the planned evidence.
For example, interviews with 20 students may reveal how they experience AI-supported learning, but they generally cannot establish the proportion of all university students who experience a particular problem.
Similarly, a qualitative account of perceived learning benefits does not ordinarily estimate the average causal effect of an intervention on examination performance.
Qualitative research may also be poorly matched when the proposed data sources cannot capture the relevant practices or when the selected methodology is inconsistent with the intended conclusion.
The problem is not the absence of numerical analysis. It is a mismatch between the question and the evidence.
If the research primarily requires population estimates, numerical comparisons, or predictive performance measures, a quantitative approach may be more appropriate.
When Should Qualitative Research Be Combined With Quantitative Evidence?
Mixed methods may be useful when the research question requires complementary forms of evidence that neither approach can adequately provide alone.
For example, a university might want to estimate how common distrust of AI-generated feedback is among students and investigate how students develop judgments about that feedback.
A quantitative survey could estimate the distribution of predefined trust-related measures, while qualitative inquiry could examine the reasoning and experiences associated with those judgments.
However, the researcher must specify how the two components will be related and what their integration is intended to establish.
Adding a small set of interviews to a survey does not automatically constitute a coherent mixed-methods design.
Likewise, including descriptive statistics in a qualitative study does not necessarily change its overall methodological orientation.
The appropriate decision depends on whether both forms of evidence are genuinely required by the research question.
Watch Out
Do not choose qualitative research simply because the available sample is small, statistical analysis seems difficult, or the topic appears subjective. A qualitative approach requires a coherent research purpose, suitable evidence, and an analytical strategy capable of supporting the intended interpretation or explanation.
