Post-Positivism

Post-positivism (also written postpositivism) is a research philosophy which holds that a real world exists independently of the researcher, but that our knowledge of it is always imperfect, provisional and open to revision. It keeps the scientific aims of positivism, such as measurement, hypothesis testing and the search for causes, while giving up positivism’s claim that careful observation can deliver certain, value-free truth.

For a dissertation, the practical question is not simply “what is post-positivism?” It is:

Do my assumptions really fit post-positivism rather than positivism, critical realism or pragmatism, and how do I justify that choice in my methodology chapter?

This page explains what post-positivism assumes, how it differs from positivism and critical realism, why the term means different things in different textbooks, and what a post-positivist dissertation looks like in practice. It ends with wording you can adapt for the philosophy section of your methodology chapter. For the wider picture, see the overview of research philosophy.

On this page:

Post-Positivism Explained Simply

Imagine a doctor reading a blood test. The test measures something real: the level of a substance in the patient’s blood. But the doctor knows that the laboratory equipment has a margin of error, that the result can vary with the time of day, and that one reading never settles a diagnosis. So the doctor treats the result as strong but fallible evidence, orders a second test if the stakes are high, and stays ready to change the diagnosis if new evidence appears.

That is the post-positivist attitude to research. Reality is there to be studied, and some explanations of it are better supported than others. But every measurement contains error, every researcher brings expectations, and every finding is the best current estimate rather than a final proof. The response is not to give up on objectivity, but to work harder at it: use several measures, test hypotheses that could fail, look for rival explanations, expose the work to critical review and report the limits honestly.

Historically, post-positivism grew out of the criticisms of logical positivism in the twentieth century. Popper (1959) argued that scientific theories can never be proven true by observation; they can only be tested and, if they fail, falsified. Kuhn (1962) showed that scientists work within shared paradigms that shape which problems and which evidence count. Post-positivism accepts these lessons without concluding that all knowledge is merely opinion. Phillips and Burbules (2000) present it as a position that recognises the limits of research while still pursuing claims that are warranted by evidence and argument.

Dudovskiy Research Assistant

Not sure whether post-positivism fits your dissertation?

Enter your topic to receive a free, tailored methodology preview.

Free · No account required
Learn more about Dudovskiy Research Assistant

The Core Assumptions of Post-Positivism

A research philosophy is defined by its assumptions about reality (ontology), knowledge (epistemology), values (axiology) and method. Guba and Lincoln (1994) summarised post-positivism along these lines, and their summary is still the one most often cited in methodology chapters.

Dimension Post-positivist assumption What it means for your study
Ontology A real world exists, but it can only be understood imperfectly (often called critical realist ontology) You study something that exists beyond your own interpretation, such as employee turnover or customer loyalty, while accepting that your account of it is incomplete
Epistemology Objectivity is an ideal to strive for, not a guarantee; findings are “probably true” and remain open to revision You claim support for, not proof of, your hypotheses, and you explain how you reduced bias
Axiology Researchers’ values can influence research, so they must be controlled and disclosed rather than assumed away You acknowledge your expectations, pre-specify hypotheses and analyses where possible, and invite scrutiny
Methodology Modified experimental and survey methods; testing hypotheses so they could be falsified; several methods and measures to cross-check findings; qualitative methods may be included You design tests that could fail, use more than one measure where possible, and look for rival explanations

Guba and Lincoln (1994) describe the post-positivist methodology as “critical multiplism”: approaching the same question from several angles so that the weaknesses of one method are offset by the strengths of another. In a dissertation, that can mean combining a survey with organisational records, or following a statistical test with interviews that probe why the result occurred.

Creswell and Creswell (2018) give a practical summary that many students meet first. They describe the postpositivist worldview as the traditional form of research, associated mostly with quantitative studies, and characterise it by determinism (causes probably determine effects), reductionism (complex ideas are broken into testable variables), empirical observation and measurement, and theory verification.

Post-Positivism vs Positivism

The two positions share a commitment to an external reality, empirical evidence and systematic method. The difference lies in how much confidence they place in what research can deliver.

Question Positivism Post-positivism
Can we know reality? Yes, through careful observation and measurement Only partly and imperfectly; knowledge is an approximation
What does a confirmed hypothesis show? That the theory is verified That the theory has survived a test it could have failed, for now
Is the researcher neutral? The researcher can stand apart from what is studied Bias is always possible and must be reduced, disclosed and checked by others
Are observations pure facts? Observation gives direct access to facts Observations are shaped by theories, instruments and expectations, so they contain error
Role of qualitative data Usually excluded as unscientific Can be included, especially to explain or test findings
How are findings stated? As facts or laws As probable, conditional and revisable conclusions

In practice, very few researchers today hold the strong positivist position in the left-hand column. Clark (1998) argued that much criticism of quantitative research attacks an outdated positivism that few researchers still defend, and that post-positivism is the more accurate description of how empirical research is actually conducted. That has a direct consequence for your dissertation: if your study tests hypotheses but you also discuss measurement error, limitations and the provisional nature of your findings, you are already reasoning as a post-positivist, even if your textbook uses the word positivism.

Post-Positivism, Critical Realism and Interpretivism

Critical realism is the position most easily confused with post-positivism, and the boundary is genuinely blurred: Guba and Lincoln (1994) describe post-positivist ontology itself as critical realist. The difference is mainly one of emphasis. Post-positivism focuses on testing hypotheses about observable relationships between variables and on reducing error in measurement. Critical realism, discussed on the realism page, focuses on explaining the underlying structures and mechanisms that produce what we observe, many of which cannot be measured directly, and it gives more weight to context and to the possibility that the same mechanism produces different outcomes in different settings.

A simple test: if your research question asks whether and how strongly X affects Y, post-positivism fits naturally. If it asks why X affects Y in some organisations and not in others, and you expect the answer to lie in structures and mechanisms that are not directly observable, critical realism is likely to fit better.

Interpretivism differs more sharply. It treats social reality as constructed through people’s meanings and interpretations, so the aim is understanding rather than measurement and causal explanation. A post-positivist asks how far flexible working changes engagement scores; an interpretivist asks what flexible working means to the employees who experience it. Pragmatism, finally, puts the research question first and chooses whatever methods answer it best, which is why it is often used to justify mixed methods research.

Two Meanings of “Post-Positivist”

One reason students find this topic confusing is that the term is used in two quite different ways, and you need to know which one your sources mean.

Meaning 1: a refined positivism. This is the sense used by Phillips and Burbules (2000), Guba and Lincoln (1994) and Creswell and Creswell (2018), and the sense used on this page. Post-positivism continues the scientific project of explanation, measurement and hypothesis testing, but with fallible knowledge and critical scrutiny built in. It is closely associated with quantitative and mixed methods research.

Meaning 2: anything that comes after positivism. Some authors use “post-positivist” for the broad family of approaches that reject positivism altogether. Ryan (2006), in a thesis-writing guide for postgraduate students, uses the term for research that emphasises meaning, reflexivity, discourse, power and narrative, and that can support social justice aims. Crossan (2003) similarly associates post-positivist thinking with qualitative exploration of phenomena in depth. In this usage, “post-positivist” sits much closer to interpretivism and critical theory.

Both usages exist in the published literature, so neither is simply wrong. The problem arises when a dissertation cites a source using one meaning to justify a design built on the other: for example, citing Ryan (2006) to support a hypothesis-testing survey. To avoid this, define the term in one sentence when you first use it, and cite sources that use it in the same sense. Henderson (2011) shows how a field can be divided on the point: writing about leisure studies, she argued that post-positivism deserved recognition alongside the familiar positivist and interpretivist paradigms, and that it could accommodate pragmatic, partly qualitative ways of examining complex meanings.

A related source of confusion comes from textbooks. Creswell and Creswell (2018) list four worldviews, starting with postpositivism. Saunders, Lewis and Thornhill (2019), widely used in UK business schools, present five philosophies (positivism, critical realism, interpretivism, postmodernism and pragmatism) and do not give post-positivism its own heading. If your course uses Saunders and your supervisor expects its categories, a design that a Creswell reader would call post-positivist will usually be described there as positivism with acknowledged limitations, or as critical realism, depending on its emphasis. Whichever label you use, what examiners reward is a clear statement of your assumptions and a design that follows from them.

Is Post-Positivism the Right Choice for Your Dissertation?

The diagram below sets out three questions that separate post-positivism from its neighbours. It is a heuristic for thinking, not a rule: real studies sit on continua, and your supervisor’s preferred terminology matters.

Decision diagram: is post-positivism the right label for your study? Three questions about reality, certainty and the main aim of the study, leading to interpretivism, positivism, post-positivism, critical realism or pragmatism

Post-positivism is usually a good fit when most of the following are true:

  • Your research questions ask about relationships, effects or differences that can be measured, and you can state hypotheses before collecting data.
  • You follow a deductive approach, moving from existing theory to tests of it.
  • You expect your instruments to contain error and plan to deal with it, for example by checking reliability and construct validity.
  • You intend to discuss rival explanations and the limits of your findings rather than present them as proof.
  • You may add qualitative data, but mainly to explain, test or illustrate quantitative findings.

It is a poor fit when your aim is to understand how participants themselves construct meaning, when you are not testing any prior theory, or when your study is designed to change a situation together with participants, as in participatory action research.

Worked Example: A Post-Positivist Dissertation

Consider a master’s dissertation asking: Does access to flexible working increase employee engagement in mid-sized professional services firms? The table compares how a naive positivist and a post-positivist version of the same study would be designed and reported.

Design decision Naive positivist version Post-positivist version
Hypothesis “Flexible working increases engagement.” H1 stated before data collection, with the result that would count against it (no difference, or lower engagement among flexible workers)
Measurement One survey question on engagement An established multi-item engagement scale, with its reliability checked in this sample, plus absence records from HR as a second, independent indicator
Rival explanations Not discussed Selection (more engaged employees may be the ones who ask for flexible working), job level and manager quality measured and controlled
Analysis A significant p-value reported as proof Effect size and confidence interval reported; result described as support for H1 under the conditions studied
Qualitative element None A small number of follow-up interviews to explore how flexible working might affect engagement and whether the survey missed anything
Conclusion “Flexible working causes higher engagement.” “The findings are consistent with a positive effect of moderate size; because the design is cross-sectional, reverse causation cannot be ruled out, and the result may not hold in other sectors.”

Notice that the post-positivist version is not less ambitious. It still aims at a causal claim about the real world. It simply builds the possibility of being wrong into every stage, which makes the eventual claim more credible to an examiner.

What to Write in Your Methodology Chapter

The philosophy section of a methodology chapter should do three things: name the position, state its assumptions in your own words, and show how those assumptions shaped your design. The template below follows that order. Replace the bracketed parts with the details of your study, and only cite sources you have read.

This study adopts a post-positivist research philosophy. It assumes that [the phenomenon, e.g. employee engagement] exists independently of the researcher and can be studied empirically, but that any knowledge of it is imperfect and open to revision (Phillips and Burbules, 2000). Consistent with this position, the study does not claim to prove its hypotheses; it tests them in a way that allows them to be contradicted by the evidence, following Popper’s (1959) principle of falsification. Because measurement is assumed to contain error, [engagement] is measured with an established multi-item scale whose reliability is assessed in this sample, and findings are cross-checked against [a second source, e.g. HR absence records], in line with the use of multiple methods and measures associated with post-positivism (Guba and Lincoln, 1994). Rival explanations, particularly [name them], are addressed through [control variables / design choices]. A positivist position was rejected because it assumes that observation can establish findings with certainty; an interpretivist position was rejected because the study aims to test relationships between measured variables rather than to explore participants’ meanings.

The last sentence matters. Examiners look for evidence that you considered alternatives and rejected them for reasons connected to your research question, not because one philosophy is “better” in general.

Post-Positivism in Business Research

A large share of quantitative business and management research is post-positivist in practice, even when it does not use the word. Survey studies that test hypotheses about, for example, the effect of service quality on customer loyalty, or of leadership style on staff turnover, typically use validated scales, report reliability and validity checks, control for alternative explanations and discuss limitations. Those are post-positivist habits.

The label question is sharper in business schools than elsewhere because of the textbook differences described above. If your programme follows Saunders, Lewis and Thornhill (2019), check with your supervisor before introducing post-positivism as a separate heading. Two approaches are usually acceptable. You can adopt positivism, as the textbook frames it, and explicitly acknowledge post-positivist qualifications (fallible measurement, provisional findings). Or you can adopt post-positivism, define it with sources such as Phillips and Burbules (2000) or Guba and Lincoln (1994), and explain briefly why you prefer it to the textbook’s categories. What weakens a business dissertation is not the choice of label but a mismatch: claiming positivism and then interpreting interviews thematically, or claiming post-positivism without any hypotheses or measurement.

Common Mistakes

  • Treating post-positivism as “positivism plus qualitative data”. The key difference is the stance towards knowledge (fallible, provisional, open to criticism), not the presence of interviews.
  • Citing sources that use the other meaning. A source that uses “post-positivist” for reflexive, meaning-centred research does not support a hypothesis-testing design.
  • Claiming post-positivism but writing as a strict positivist. Phrases such as “the results prove” or “the data show conclusively” contradict the philosophy you declared.
  • Naming a philosophy that does nothing. If the philosophy section could be deleted without changing any design decision, it has not been applied. Link it explicitly to your hypotheses, measures and limitations.
  • Confusing post-positivism with postmodernism. Postmodernism questions the possibility of objective knowledge and grand explanations; post-positivism keeps objectivity as an aim.
  • Using hypotheses without a real test. A hypothesis that no possible result could contradict is not doing post-positivist work.

Post-Positivism in the Age of AI and Digital Research

Digital data and AI tools have revived an old temptation: the belief that, with enough data, findings simply emerge without theory or interpretation. Kitchin (2014) examined claims that big data would bring “the end of theory” and argued that the epistemological assumptions of data-driven research need critical reflection rather than acceptance. Post-positivism offers a useful framework for that reflection. Platform data, web-scraped text and online behaviour records are measurements produced by particular systems for particular purposes. They contain errors and biases, just as questionnaires do, and their size does not remove them.

The same applies when AI tools are used inside a study, for example to classify open-ended survey answers, extract variables from documents or summarise literature. From a post-positivist standpoint, an AI classification is a measurement with unknown error until it has been checked. A defensible dissertation reports how the tool was used, compares a sample of its output against human coding (see inter-rater reliability), and treats disagreements as information about measurement quality. Likewise, AI-generated literature summaries are claims to be verified against the original sources, not evidence in themselves.

Post-positivism’s emphasis on falsifiable hypotheses also helps when large datasets make it easy to find patterns by chance. Stating hypotheses and analysis plans before exploring the data, and labelling any later discoveries as exploratory, keeps a digital dissertation honest about what was tested and what was found along the way.

Advantages and Limitations

Advantages. Post-positivism matches how most contemporary empirical research is actually done, so it is easy to defend. It supports clear, testable research questions and a coherent link between philosophy, hypotheses and methods. Its honesty about error and bias encourages good practice in measurement and reporting, and it allows qualitative evidence to play a supporting role, which makes it a natural basis for some mixed methods designs.

Limitations. Critics from interpretivist and critical traditions argue that post-positivism still privileges measurement and can reduce complex human experience to variables. Its ontology overlaps with critical realism, so the distinction can look thin to examiners. The term’s double meaning creates confusion if it is not defined. And in programmes built around textbooks that do not use the term, choosing it may require extra justification.

Exam Tip

If you are asked to compare positivism and post-positivism, avoid presenting them as opposites. The strongest answers explain that post-positivism keeps positivism’s commitment to an external reality and empirical testing, but replaces verification with falsification and certainty with fallible, probable knowledge. Name the sources of that shift (Popper’s falsification and Kuhn’s paradigms), give one concrete design consequence, such as using several measures or reporting confidence intervals rather than claiming proof, and note that the term is sometimes used more broadly for any post-positivist alternative. That combination shows understanding rather than memorised definitions.

Not sure whether post-positivism, positivism or critical realism best fits your study? Dudovskiy Research Assistant can review your research questions and planned design, suggest a research philosophy that is consistent with them and help you word its justification in your methodology chapter. It supports your own judgement; the final decisions remain yours and your supervisor’s.

John Dudovskiy

References

Clark, A.M. (1998). The qualitative-quantitative debate: moving from positivism and confrontation to post-positivism and reconciliation. Journal of Advanced Nursing, 27(6), 1242–1249. https://doi.org/10.1046/j.1365-2648.1998.00651.x

Creswell, J.W. and Creswell, J.D. (2018). Research Design: Qualitative, Quantitative, and Mixed Methods Approaches. 5th edn. Thousand Oaks, CA: SAGE.

Crossan, F. (2003). Research philosophy: towards an understanding. Nurse Researcher, 11(1), 46–55. https://doi.org/10.7748/nr2003.10.11.1.46.c5914

Guba, E.G. and Lincoln, Y.S. (1994). Competing paradigms in qualitative research. In: N.K. Denzin and Y.S. Lincoln (eds.), Handbook of Qualitative Research. Thousand Oaks, CA: SAGE, pp. 105–117.

Henderson, K.A. (2011). Post-positivism and the pragmatics of leisure research. Leisure Sciences, 33(4), 341–346. https://doi.org/10.1080/01490400.2011.583166

Kitchin, R. (2014). Big Data, new epistemologies and paradigm shifts. Big Data & Society, 1(1), 1–12. https://doi.org/10.1177/2053951714528481

Kuhn, T.S. (1962). The Structure of Scientific Revolutions. Chicago: University of Chicago Press.

Phillips, D.C. and Burbules, N.C. (2000). Postpositivism and Educational Research. Lanham, MD: Rowman & Littlefield.

Popper, K.R. (1959). The Logic of Scientific Discovery. London: Hutchinson.

Ryan, A.B. (2006). Post-positivist approaches to research. In: Researching and Writing your Thesis: A Guide for Postgraduate Students. Maynooth: MACE, pp. 12–26.

Saunders, M.N.K., Lewis, P. and Thornhill, A. (2019). Research Methods for Business Students. 8th edn. Harlow: Pearson.

Last reviewed: November 2026.

[]