Qualitative Paradigms and Research Questions

Overview

A research paradigm is a broad set of assumptions about what can be known, how knowledge is produced and what counts as a defensible explanation. Paradigms matter in qualitative research because the researcher is not a neutral measuring device positioned outside the phenomenon. Choices about whose perspective matters, what constitutes evidence, how language is interpreted and what kind of claim is made are shaped by underlying assumptions.[1,2]

Health researchers do not need to turn every protocol into a philosophical essay. They do, however, need enough conceptual clarity to avoid contradictory designs—for example, claiming that experiences are socially constructed while analysing interviews as if each statement were a direct, context-free measurement of one objective reality.

This lecture explains ontology, epistemology and common paradigmatic orientations in practical terms and shows how they shape qualitative questions, methods and interpretation.

Why this matters

Two researchers can interview the same participants and legitimately ask different questions of the data. One may seek a close description of participants’ experiences. Another may examine how professional language constructs categories such as “adherent” or “non-adherent”. A third may ask how power and institutional structures shape whose account is treated as authoritative.

The difference is not merely analytic technique. It reflects different assumptions about reality, knowledge and explanation. Qualitative quality therefore depends partly on coherence: the question, theoretical orientation, data-generation method, analysis and claims should fit together.[2-4]

Key concepts

Ontology: what kind of reality is assumed?

Ontology concerns assumptions about reality. A more realist position assumes that important features of the world exist independently of individual interpretations, even if they are imperfectly known. A relativist or constructivist position emphasises that some social realities are produced through historical, cultural and relational processes and may differ legitimately across people or groups.[1]

In health research, this distinction need not be extreme. A biomedical event such as a laboratory value can be materially real while its meaning—whether it is experienced as threat, stigma, reassurance or irrelevant information—may be socially and personally constructed.

Epistemology: how can we know about it?

Epistemology concerns the relationship between the knower and what is being studied. In qualitative research, knowledge is often understood as situated: what participants say depends partly on context, language, memory, relationships and the interview situation. The researcher’s interpretation is also shaped by disciplinary training and prior assumptions.[2]

Methodology: how should inquiry proceed?

Methodology is the organised logic connecting philosophical assumptions to the research process. It guides choices such as phenomenology, grounded theory, ethnography or case study. Methods are the practical procedures—interviews, observation, documents, coding and analytic techniques.

Common paradigmatic orientations

Postpositivist or subtle realist orientations

A postpositivist orientation assumes that an external reality exists but can only be known imperfectly. Qualitative methods may be used to obtain participants’ accounts while researchers use systematic comparison, negative cases and transparent procedures to develop credible explanations. Health-services studies often adopt an implicit realist orientation even when they do not use the term.[2,3]

Questions may ask: What barriers and facilitators influence implementation of the postpartum pathway? The researcher treats accounts as evidence about real processes while acknowledging incomplete access to those processes.

Constructivist or interpretivist orientations

Constructivist approaches emphasise that meanings are developed through interaction, culture and context. The researcher does not assume that one account reveals an unmediated truth hidden inside the participant. Instead, the interview itself is part of the knowledge-producing situation.[1]

Questions may ask: How do women make sense of future diabetes risk after pregnancy? Multiple interpretations can coexist and may change across time and relationships.

Critical orientations

Critical approaches examine how power, institutions, social categories and material inequalities shape experience and knowledge. They often ask whose perspective is privileged, which assumptions appear natural, and how research may reproduce or challenge inequity.

A critical question might ask: How do clinic policies and gendered expectations distribute responsibility for postpartum follow-up between women, families and health services?

Pragmatic orientations

Pragmatism is often associated with mixed methods, but qualitative researchers may also work pragmatically by selecting concepts and methods according to the practical problem while remaining explicit about assumptions. Pragmatism does not mean “anything goes”. The methods must still be coherent with the question and capable of supporting the intended conclusion.

How paradigm can change the research question

illustrative

Table 1. Different lenses on one health-service problem

Orientation	Illustrative question	What becomes analytically important?
Postpositivist / realist	What barriers and facilitators influence test completion?	Processes that plausibly help explain behaviour
Constructivist	How do women make sense of postpartum diabetes risk and follow-up?	Meaning, interpretation and context
Critical	How do institutional arrangements shape responsibility and access?	Power, resources and taken-for-granted norms
Pragmatic	What experiences should inform redesign of the follow-up pathway?	Actionable knowledge relevant to the service problem

Formulating qualitative research questions

Good qualitative questions are open enough to permit discovery but focused enough to guide sampling and data generation. They usually identify:

• the phenomenon or process;
• the participant group or social setting;
• the form of understanding sought; and
• sometimes the context or stage of care.

DECISION GUIDE

Test the coherence of a qualitative question

1. Does the question seek meaning, process, interaction, experience or contextual explanation?
2. Does the wording prematurely assume a cause, deficit or moral judgement?
3. Is the phenomenon narrow enough for intensive qualitative inquiry?
4. Does the chosen paradigm help answer the question rather than merely decorate the protocol?
5. Will the proposed participants, setting and data sources provide access to the phenomenon?
6. Does the planned analytic approach support the type of claim implied by the question?

Compare “Why are women non-compliant?” with “How do women describe decisions about postpartum glucose testing?” The first embeds a judgement and may assume that failure lies primarily with the patient. The second opens space for competing explanations involving service design, family context, beliefs and practical constraints.

Theory in qualitative research

Theory can enter qualitative inquiry in different ways. It may sensitise researchers to concepts, guide sampling, support interpretation or be developed from the data. Reeves and colleagues emphasise that theory can make qualitative analysis more explanatory by connecting observations to broader concepts.[2]

However, theory can also narrow attention prematurely. A study using the Health Belief Model, for example, should not force every interview statement into predetermined domains if the purpose is exploratory and participants raise organisational or relational issues outside the framework.

PROTOCOL CHECKLIST

Specify the conceptual position without overloading the protocol

• State the qualitative methodological orientation.
• Briefly describe the epistemological stance if it materially affects interpretation.
• Name any formal theory used and explain its role.
• Describe the researcher’s relationship to participants and setting.
• Explain whether coding will be primarily inductive, deductive or iterative between both.
• State the form of knowledge claim the study intends to make.

Reflexivity and positionality

Paradigms affect how researcher influence is treated. In many qualitative approaches, reflexivity is preferable to pretending that the researcher has no effect. A clinician may notice subtleties that an outsider misses but may also take familiar practices for granted. A researcher from outside the service may invite criticism more readily but misinterpret local shorthand.

The methodological response is to document positionality, discuss assumptions within the research team, use reflexive memos and examine how interpretations were reached. Reflexivity strengthens transparency; it does not guarantee immunity from bias.[3-5]

Ethical implications of paradigmatic choices

Questions that expose structural inequality or sensitive identities may also create risks if quotations are identifiable. Researchers should anticipate how interpretation and dissemination might affect the communities being described.

Worked example

WORKED EXAMPLE

Reframing postpartum follow-up through different paradigms

The research team begins with a deficit-oriented question: “Why are women non-adherent with postpartum testing?” During protocol review, they recognise that the wording presumes a patient failure.

For a constructivist interview study they reformulate the question as: How do women make sense of postpartum diabetes risk and the recommendation for glucose testing after pregnancy? Sampling seeks contrasting experiences, and interviews examine language, expectations and transitions in identity from “high-risk pregnancy patient” to “healthy mother”. Analysis will focus on patterned meanings rather than counting barriers.

For a critical service study, the question becomes: How do organisational arrangements and professional expectations shape responsibility for postpartum diabetes prevention? Data would include women, clinicians and relevant service documents. Analysis would examine how responsibility is distributed and how resource constraints affect whose explanations dominate.

Both are coherent qualitative projects, but they are not interchangeable. Their questions, data sources and claims differ because their ways of knowing differ.

Suggested protocol or reporting wording

How paradigms change the meaning of the same question

Paradigms matter most when they change what the researcher treats as a legitimate answer. Consider the question, “Why do women not complete postpartum glucose testing?” A post-positivist study may seek explanatory factors that can be measured and related to non-completion. A constructivist qualitative study may ask how women and clinicians construct the meaning and priority of follow-up after pregnancy. A critical inquiry may ask how institutional arrangements, gendered care responsibilities or unequal access distribute the burden of follow-up. A pragmatic mixed-methods study may combine these questions because the service decision requires both pattern estimation and contextual explanation.[1-3]

These versions are not merely different wordings. They define different objects of knowledge. The researcher should therefore avoid selecting a paradigm after the methods have already been fixed simply to make the protocol appear philosophically sophisticated.

Researcher-participant relationships

Different paradigmatic positions also imply different understandings of the research relationship. In many interpretive traditions, interview data are viewed as co-produced through interaction. What a participant says may depend on who asks, how the question is framed, what the participant believes the interviewer wants to hear and what can safely be discussed in that relationship.[3,4]

This does not mean that qualitative findings are arbitrary. It means the researcher should describe and reflect on the conditions under which knowledge was generated. A clinician interviewing her own patients about dissatisfaction with care faces a different relational context from an independent researcher interviewing the same participants.

Theory can sensitise without determining

A theoretical framework can provide concepts that sensitise the researcher to processes that might otherwise be overlooked.[2] For example, continuity-of-care theory may draw attention to informational, relational and management continuity. However, the framework should not become a filter that makes every account fit predetermined concepts. Researchers can explicitly state whether theory informed question development, sampling, interview prompts, coding, interpretation or only the discussion.

Pragmatism and mixed-methods questions

Pragmatism is often associated with mixed-methods research because it prioritises methods that help answer a practical research problem. It should not be reduced to “anything goes”. A pragmatic design still requires coherent questions, defensible methods and explicit integration. Part 12 develops this in detail. At this stage, the important principle is that the research question should determine whether one form of evidence is sufficient or whether several forms of knowledge are needed.

Appraising paradigm–question coherence

When reviewing a qualitative proposal, ask whether the stated paradigm has observable consequences. If a study claims a constructivist orientation but uses a rigid interviewer script, treats coding as objective classification and reports one definitive truth without reflexivity, the philosophical label and practical design are poorly aligned. Conversely, a study does not need pages of philosophy if its assumptions are already clear through a coherent question, methodology and analytic approach.[4,5]

A concise protocol can therefore explain: what kind of knowledge is sought, how the researcher understands the relationship between participant accounts and interpretation, and how those assumptions shaped design choices. That is more useful than listing ontology and epistemology terms without application.

A practical test of coherence

Before finalising the protocol, restate the question, intended knowledge claim, participant relationship and analytic product in ordinary language. If these cannot be explained without relying on philosophical labels, the design may not yet be coherent. A paradigm should clarify why the chosen evidence and interpretation are appropriate, not make a straightforward study unnecessarily obscure.

Common mistakes

Knowledge check

What is the practical difference between methodology and method?

Answer and reasoning

Methodology is the organising logic of inquiry that connects assumptions, question and analytic strategy. Methods are the specific procedures, such as interviews, observation or coding. The same method can be used within different methodologies for different purposes.

Why might 'Why are patients non-compliant?' be a poor qualitative research question?

Answer and reasoning

It embeds a judgement and assumes that the problem is located in patients before the inquiry begins. A more open question allows patient, organisational, relational and contextual explanations to emerge.

Can a constructivist researcher use an interview guide?

Answer and reasoning

Yes. A semi-structured guide is compatible with constructivist inquiry if it supports the research question and the researcher remains responsive to how meanings are produced in interaction. Constructivism does not require completely unstructured conversation.

What should a reader look for when a paper claims a critical qualitative orientation?

Answer and reasoning

The research question, sampling, analysis and interpretation should actually engage with power, structure or inequity. The label alone is insufficient if the study merely lists individual barriers without examining how social arrangements shape them.

Key take-home messages

SUMMARY

Key take-home messages

• Paradigms influence what researchers treat as reality, evidence and explanation.
• Use philosophical language only when it clarifies practical methodological choices.
• Qualitative questions should be open, non-judgemental and aligned with the intended form of knowledge.
• Theory can deepen interpretation but should not automatically force all data into predetermined categories.
• Coherence among paradigm, question, methods, reflexivity and claims is a central marker of qualitative quality.

References

1. Guba EG, Lincoln YS. Competing paradigms in qualitative research. In: Denzin NK, Lincoln YS, editors. Handbook of qualitative research. Thousand Oaks (CA): Sage; 1994. p. 105-117.
2. Reeves S, Albert M, Kuper A, Hodges BD. Why use theories in qualitative research? BMJ. 2008;337:a949. https://doi.org/10.1136/bmj.a949
3. Malterud K. Qualitative research: standards, challenges, and guidelines. Lancet. 2001;358(9280):483-488. https://doi.org/10.1016/S0140-6736(01)05627-6
4. O’Brien BC, Harris IB, Beckman TJ, Reed DA, Cook DA. Standards for reporting qualitative research: a synthesis of recommendations. Acad Med. 2014;89(9):1245-1251. https://doi.org/10.1097/ACM.0000000000000388
5. Tracy SJ. Qualitative quality: eight “big-tent” criteria for excellent qualitative research. Qual Inq. 2010;16(10):837-851. https://doi.org/10.1177/1077800410383121