Research Paradigms and Ways of Knowing

Overview

A research paradigm is a coherent set of assumptions about what exists, how knowledge can be developed, what role values play and which forms of inquiry are appropriate. These assumptions influence the questions researchers consider meaningful, the relationship between researcher and participant, the type of evidence sought, the methods selected and the criteria used to judge quality [1].

Health researchers do not need to become professional philosophers. They do need enough philosophical awareness to avoid combining incompatible claims without explanation. A study that treats blood pressure as a measurable biological quantity raises different knowledge questions from a study exploring what living with hypertension means to patients. Both can be rigorous, but their evidence, researcher roles and quality criteria are not identical.

Why this matters

Paradigms are often hidden. Researchers may select methods by habit, then add a philosophical label near the end of a thesis. This reverses the logic. The assumptions should help explain why the question, design, data generation and interpretation fit together.

Philosophical awareness is especially important in qualitative and mixed-methods research, but quantitative research also rests on assumptions about observation, measurement, error and the extent to which reality can be known. Positivist and post-positivist assumptions have strongly influenced biomedical research [2]. Interpretivist and constructivist approaches have shaped studies of experience, meaning, context and interaction. Pragmatism is often used when the research problem requires more than one form of evidence [3].

Five connected terms

Ontology: what is assumed to exist?

Ontology concerns the nature of reality. In a health study, a researcher might assume that there is a biological state that exists independently of observation, while also recognising that experiences such as stigma or trust are shaped through social interaction and context.

Epistemology: how can knowledge be developed?

Epistemology concerns the relationship between the knower and what is known. It asks what counts as evidence and how confident knowledge claims can be. A controlled measurement may be central in one study; co-constructed accounts between participant and interviewer may be central in another.

Axiology: how do values matter?

Axiology concerns values and ethics. It includes which questions are prioritised, whose perspectives are included, how the researcher’s position influences the study and what consequences the research may have. No research is value-free in the sense that all choices disappear; the aim is to make important values and interests visible and responsibly managed.

Methodology: what logic guides the inquiry?

Methodology is the strategy connecting assumptions to the research process. Examples include experimental inquiry, epidemiological observation, phenomenology, grounded theory, ethnography and mixed-methods designs.

Methods: what specific techniques are used?

Methods include surveys, assays, interviews, observations, randomisation, record extraction, coding and statistical models. The same method may be used within different methodologies for different purposes.

Major paradigmatic perspectives

Labels vary across disciplines, and boundaries are not absolute. The following perspectives are useful introductory guides rather than rigid boxes.

Positivism

Positivism assumes that an objective reality can be observed and described through systematic measurement. The researcher seeks distance from the object of study, standardisation and control. Hypotheses are tested using observable data, and quality is associated with reliability, validity, replicability and minimisation of bias.

Strict positivism is less common in contemporary health research than simplified textbook descriptions suggest. Measurement itself can be imperfect, and observations are influenced by theory and instruments.

Post-positivism

Post-positivism retains the idea that a reality exists independently of the researcher but accepts that it can be known only imperfectly. Findings remain provisional and vulnerable to error. Researchers use falsifiable questions, multiple measurements, replication, uncertainty estimates and critical appraisal to reduce the chance of mistaken conclusions [2].

Much quantitative clinical and epidemiological research is compatible with post-positivist reasoning: it seeks estimates of real phenomena while explicitly acknowledging sampling variability, measurement error, bias and model assumptions.

Interpretivism

Interpretivism focuses on understanding how people make sense of experiences and social worlds. Meaning is situated in context rather than treated as a fixed quantity waiting to be measured. The researcher is part of the knowledge-generating process, and interpretation is central.

An interpretivist study of medication adherence may examine how patients interpret an asymptomatic condition, negotiate advice from family and clinicians, and fit treatment into daily life.

Constructivism

Constructivism emphasises that meanings and understandings are constructed through social, cultural and historical interaction. Multiple realities may coexist because people experience and interpret the same situation differently. Researcher and participant may co-construct the account produced in an interview.

Interpretivism and constructivism overlap and are sometimes used together. The important issue is not choosing the perfect label but explaining the assumptions and their consequences for method and interpretation [4].

Pragmatism

Pragmatism begins with the research problem and the practical consequences of inquiry. It allows researchers to draw on different methods and forms of evidence when doing so provides a more useful answer. Pragmatism is commonly associated with mixed-methods research, but it is not a licence to combine methods without rationale. The study must still explain why each strand is needed and how findings are integrated [3].

Introductory comparison of research paradigms

illustrative

Table 1. Broad paradigmatic tendencies

Perspective	View of knowledge	Researcher role	Common question emphasis	Common quality emphasis
Positivism	Knowledge through objective observation and measurement	Seeks detachment and control	What is the measurable relationship or effect?	Validity, reliability, objectivity, replication
Post-positivism	Reality exists but knowledge is provisional and fallible	Uses critical testing and error reduction	What explanation remains compatible with the evidence?	Bias control, uncertainty, triangulation, falsifiability
Interpretivism	Knowledge arises through interpretation of meaning in context	Engages with participants and interprets accounts	How is an experience understood?	Credibility, contextual depth, reflexivity
Constructivism	Meanings are socially and contextually constructed	Co-constructs knowledge with participants	How are realities produced and negotiated?	Coherence, reflexivity, representation of perspectives
Pragmatism	Knowledge is judged partly by its usefulness for the problem	Selects and integrates appropriate forms of inquiry	What combination of evidence best addresses the problem?	Fitness for purpose, integration, practical consequences

Note: Real studies may draw on more nuanced positions. The table is a teaching simplification.

Paradigms are not determined by data format

Numbers are often associated with post-positivist research and words with interpretivist research, but the relationship is not automatic. A qualitative content analysis can count categories. A quantitative scale may measure a socially constructed concept. An interview can be used within a realist evaluation. The key questions are what the data are assumed to represent, how they are generated and what claims will be made.

Similarly, using open-ended and closed-ended questions in one questionnaire does not automatically create mixed-methods research. Mixed methods requires a rationale for both approaches and purposeful integration of the strands.

Alignment in practice

A paradigm influences:

• the wording of the research question;
• whether hypotheses are specified;
• how participants are selected;
• whether the researcher seeks distance or relational engagement;
• what counts as data;
• how analysis is conducted;
• what quality criteria are appropriate; and
• how conclusions are expressed.

APPRAISAL CHECKLIST

Appraise paradigmatic alignment

• Are the philosophical assumptions stated or inferable from the study aim?
• Do the research question and methodology reflect the same view of knowledge?
• Is the researcher–participant relationship described consistently with the approach?
• Are quality criteria appropriate for the methodology rather than borrowed mechanically from another tradition?
• Are conclusions limited to the kind of knowledge the study can produce?

Reflexivity and positionality

In interpretive and constructivist research, reflexivity is the disciplined examination of how the researcher’s background, role, assumptions and interactions shape the research. A positionality statement may describe professional role, relationship to the setting, relevant experiences and how these were managed during data generation and interpretation.

Quantitative researchers also make value-laden choices, including which variables to measure, which thresholds to use and which outcomes matter. Reflexivity is therefore not evidence that qualitative work is “subjective and weak”; it is one way of making the knowledge-production process transparent.

Critical and transformative perspectivesMMed extension

Critical and transformative paradigms examine how power, inequality and social structures shape both health and knowledge. They may prioritise participation, emancipation or action with affected communities. These perspectives are important in health-equity, disability, gender, indigenous-health and participatory research. They should not be claimed merely because a study discusses inequality; the design should show how power and participant involvement are addressed.

Worked example

WORKED EXAMPLE

One adherence problem, three paradigmatic framings

Post-positivist framing: What patient and treatment factors are associated with objectively measured non-adherence among adults with hypertension? The study uses pre-specified variables, a defined adherence measure and regression analysis. It seeks an estimate while acknowledging measurement error and residual confounding.

Interpretivist framing: How do adults with hypertension make sense of long-term medication use in everyday life? The study uses purposive sampling and in-depth interviews, with reflexive thematic analysis. It seeks contextualised understanding rather than a population prevalence.

Pragmatic mixed-methods framing: Which patients are at greatest risk of non-adherence, and how can the observed patterns be explained to inform a feasible clinic intervention? The study first quantifies patterns, then samples contrasting participants for interviews. Integration connects measured patterns with explanations and service recommendations.

Each framing addresses a different question. None is automatically superior. Quality depends on alignment and execution.

Suggested Methods wording

Common mistakes

Knowledge check

What is the difference between ontology and epistemology?

Answer and reasoning

Ontology concerns assumptions about the nature of reality or what exists. Epistemology concerns how knowledge about that reality can be developed and justified. They are connected but answer different questions.

Does using interviews automatically make a study interpretivist?

Answer and reasoning

No. Interviews are a method. Their paradigmatic meaning depends on how the researcher understands the interview account, the researcher–participant relationship, the analytical approach and the intended claim.

Why is post-positivism often a better description than strict positivism for modern quantitative health research?

Answer and reasoning

Post-positivism accepts that a reality may exist but that measurements and explanations are fallible. It therefore emphasises uncertainty, bias control, replication and critical testing rather than assuming perfect direct observation.

A study includes a survey and five interviews but reports the findings in separate sections without connecting them. Is it necessarily mixed methods?

Answer and reasoning

No. Mixed methods requires a clear rationale for using both approaches and purposeful integration. Merely placing quantitative and qualitative components in the same project is insufficient.

Key take-home messages

SUMMARY

Key take-home messages

• Paradigms make assumptions about reality, knowledge, values and inquiry visible.
• Ontology, epistemology, axiology, methodology and methods are connected but distinct.
• Paradigms influence questions, researcher roles, evidence, quality criteria and conclusions.
• Data format alone does not determine a paradigm or methodological approach.
• The goal is defensible alignment, not attaching a fashionable philosophical label.

References

1. Brown MEL, Dueñas AN. A medical science educator’s guide to selecting a research paradigm: building a basis for better research. Med Sci Educ. 2020;30(1):545–553. https://doi.org/10.1007/s40670-019-00898-9
2. Park YS, Konge L, Artino AR Jr. The positivism paradigm of research. Acad Med. 2020;95(5):690–694. https://doi.org/10.1097/ACM.0000000000003093
3. Morgan DL. Pragmatism as a paradigm for social research. Qual Inq. 2014;20(8):1045–1053. https://doi.org/10.1177/1077800413513733
4. Carter SM, Little M. Justifying knowledge, justifying method, taking action: epistemologies, methodologies, and methods in qualitative research. Qual Health Res. 2007;17(10):1316–1328. https://doi.org/10.1177/1049732307306927